Retrieval of Medicinal Chemical Information

Retrieval of Medicinal Chemical Information W. Jeffrey Howe, EDITOR The Upjohn Company Margaret M. Milne, EDITOR Smi...

Author: W. Jeffrey Howe | Margaret M. Milne | Ann F. Pennell

27 downloads 873 Views 4MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Retrieval of Medicinal Chemical Information W. Jeffrey Howe,

EDITOR

The Upjohn Company Margaret M. Milne,

EDITOR

Smith, Kline, and French An ICI Americas, Inc.

Based on a symposium cosponsored by the Divisions of Computers i n Chemistry and Chemical Information at the 175th Meeting of the American Chemical Society, Anaheim, California, March 13-17, 1978.

ACS SYMPOSIUM SERIES 84

AMERICAN

CHEMICAL

SOCIETY

WASHINGTON, D. C. 1978

In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

Library of Congress Œ ? Data Symposium on Retrieval of Medicinal Chemical Information, Anaheim, Calif., 1978. Retrieval of medicinal chemica (ACS symposium series; 84 ISS Based on a symposium cosponsored by the Divisions of Computers in Chemistry and Chemical Information at the 175th meeting of the American Chemical Society, California, March 1978. Includes bibliographies and index. 1. Information storage and retrieval systems—Chemistry, Pharmaceutical—Congresses. 2. Chemistry, Pharmaceutical—Data processi ng—Congresses. I. Howe, William Jeffrey, 1946- . II. Milne, Margaret M., 1946- . III. Pennell, Ann R, 1946- . IV. American Chemical Society. Division of Computers in Chemistry. V. American Chemical Society. Division of Chemical Information. VI. Title. VII. Series: American Chemical Society. ACS symposium series; 84. RS421.S93 1978 615Μ9Ό2854 78-21611 ISBN 0-8412-0465-9 ACSMC 8 84 1-231 1978

Copyright © 1978 American Chemical Society All Rights Reserved. The appearance of the code at the bottom of thefirstpage of each article in this volume indicates the copyright owner's consent that reprographic copies of the article may be made for personal or internal use or for the personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc. for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to copying or transmission by any means—graphic or electronic—for any other purpose, such as for general distribution, for advertising or promotional purposes, for creating new collective works, for resale, or for information storage and retrieval systems. The citation of trade names and/or names of manufacturers in this publication is not to be construed as an endorsement or as approval by ACS of the commercial products or services referenced herein; nor should the mere reference herein to any drawing, specification, chemical process, or other data be regarded as a license or as a conveyance of any right or permission, to the holder, reader, or any other person or corporation, to manufacture, repro duce, use, or sell any patented invention or copyrighted work that may in any way be related thereto. PRINTED IN THE UNITED

STATES

OF

AMERICA

Society Library 1155 16th St. N. W. In Retrieval of Medicinal Chemical Information; Howe, W., et al.; Washington, D. C. Society: 20036Washington, DC, 1978. ACS Symposium Series; American Chemical

ACS Symposium Series Robert F. Gould, Editor

Advisory Board Kenneth B. Bischoff

Nina I. McClelland

Donald G. Crosby

John B. Pfeiffer

Jeremiah P. Freeman

Joseph V. Rodricks

E. Desmond Goddard

F. Sherwood Rowland

Jack Halpern

Alan C. Sartorelli

Robert A. Hofstader

Raymond B. Seymour

James P. Lodge

Roy L. Whistler

John L. Margrave

Aaron Wold


FOREWORD The ACS SYMPOSIUM SERIES was founded in 1974 to provide a medium for publishin format of the Series parallels that of the continuing ADVANCES IN CHEMISTRY SERIES except that in order to save time the papers are not typeset but are reproduced as they are submitted by the authors in camera-ready form. Papers are reviewed under the supervision of the Editors with the assistance of the Series Advisory Board and are selected to maintain the integrity of the symposia; however, verbatim reproductions of previously published papers are not accepted. Both reviews and reports of research are acceptable since symposia may embrace both types of presentation.


PREFACE '""phe symposium on retrieval of medicinal chemical information was organized to examine current developments in the storage, retrieval, and manipulation of the variety of types of data that are associated with medicinal chemistry in the pharmaceutical industry, government agencies, and related organizations. This volume contains expanded versions of the papers presented at the symposium as well as several additional invited papers. To insure adequate coverage of what has become a broad and increasingly important field, the speakers were selected to approach the topic from a number of differen the manipulation of biological data, chemical substructure searching, computer graphical display of retrieved data, the integration of biological search results with chemical information, the utilization of retrieval systems in the research function, mathematical analyses of chemical-structure data bases, and so on. While some of the chapters deal with commercially available information systems, most focus on the capabilities of systems that were developed within individual organizations. Authors were encouraged to include not only what has been done in the area, but what is now being planned for implementation in the near future to meet the growing information needs of medicinal chemical research. Several papers have been included that were not presented at the symposium. These invited submissions extend the treatment of the subject beyond the limitations of a one-day symposium. However, the breadth of thefieldof medicinal chemical information has made it impossible to offer complete coverage in a volume of this size. For example, the storage and retrieval of clinical test data is one important area which could not be dealt with here. Early in the planning stages of the symposium it was recognized that some form of overview of medicinal chemical information would be a valuable addition to a proceedings volume. To that end we have written an introductory chapter, based on the contents of the symposium presentations and on discussions with symposium participants. It is hoped that this overview will do three things: (a) provide a fairly complete statement of the current status and directions of progress of the field as a whole, (b) illustrate the interrelationship of the various categories of medicinal chemical information, including those important areas which could not be dealt with in the symposium, and (c) provide a conceptual framework for viewing the material discussed in the papers which follow. A

vii In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

One other point deserves mention. The ACS Books Department has instituted a policy of peei review for papers included in the ACS Symposium Series. Although this adds time and effort to the publication process, it can only result in an improvement in the quality of the papers, and will benefit both readers and authors. We fully support this move. We wish to thank the officers of the Divisions of Chemical Information and Computers in Chemistry for their assistance, especially Gabrielle Revesz, Mary Reslock, and Ed Olson. The Upjohn Company Kalamazoo, MI 49001

W.

Smith, Kline, and French

JEFFREY HOWE

MARGARET M. MILNE

Philadelphia, PA 19101 ICI Americas, Incorporated Wilmington, DE 19897

ANN F. PENNELL

August 22, 1978

viii In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

1

R e t r i e v a l of M e d i c i n a l C h e m i c a l I n f o r m a t i o n — a n Overview MARGARET M. MILNE—Smith, Kline, and French, Philadelphia, PA 19101 ANN F. PENNELL—ICI Americas, Inc., Wilmington, DE 19897 W. JEFFREY HOWE—The Upjohn Company, Kalamazoo, MI 49001

The following pape writte by the organizers of the ACS Symposium on Retrieval based in part on the Symposium presentations, on discussions with sympusium paticipants, and on the author's own involvement in the field of pharmaceutical research and development. Traditionally, the term "medicinal chemistry" has connoted an area of synthetic organic chemistry which deals with the prepara tion of molecules likely to have some desired physiological re sponse. Associated with each synthesized molecule is a collection of in vivo or in v i t r o test results used to ascertain the actual nature and extent of the bioactivity (if any). Evolving in par a l l e l with this view, medicinal chemical information systems have commonly been based on a data f i l e organized by compound and have contained such items as chemical structure, identification number, source, and sometimes physical properties. The biological test results were also organized by compound, but either because of their volume or for administrative reasons the results were usual ly separate from the structural data f i l e . Over the past 15 years considerable effort has been invested in computerizing these f i l e s , in developing efficient, powerful, and rapid mechanisms for selective r e t r i e v a l , and in integrating the searching of struc tural data with that of biological data without actually combining the individual f i l e s . More recently, however, i t has been recognized that the t r a ditional view of medicinal chemistry is actually one element of a much larger set of functions in the total drug development proc ess. These functions are interrelated and at times interdepend ent, and the drug development process can be made markedly more effective by f a c i l i t a t i n g the necessary interactions. What this implies to developers and users of medicinal chemical information systems is a need for access to a considerably more diverse set of drug-related information types and for additional capabilities in retrieving, correlating, and displaying these data. This, in fact, is the direction of current progress in the field of medici0-8412-0465-9/78/47-084-001$05.00 © 1978 American Chemical Society In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

2

RETRIEVAL OF MEDICINAL CHEMICAL INFORMATION

n a l chemical i n f o r m a t i o n . To examine t h i s progress i n more det a i l , the i n d i v i d u a l f u n c t i o n s that comprise the t o t a l drug development process and the corresponding i n f o r m a t i o n needs w i l l now be considered. The Drug Development Process Figure 1 i l l u s t r a t e s s c h e m a t i c a l l y a h y p o t h e t i c a l drug development program. The diagram i s approximate s i n c e the sequence of f u n c t i o n s may vary somewhat between d i f f e r e n t o r g a n i z a t i o n s and s i n c e not a l l of the conceivable i n t e r a c t i o n s among f u n c t i o n s are i n d i c a t e d . I n a d d i t i o n , some of the f u n c t i o n s may overlap (e.g., p a t h o l o g y / t o x i c o l o g y s t u d i e s may continue d u r i n g c l i n i c a l t r i a l s ) and some may be done i n segments i n t e r l e a v e d among the other f u n c t i o n s (e.g., a p p l i c a t i o n s f o r Food and Drug A d m i n i s t r a t i o n (FDA) approval ma variou diagram does i n c l u d e a l imate r e l a t i o n s h i p s . Development of a drug begins w i t h s e l e c t i o n of a prototype or lead compound or compound s e r i e s whose a c t i v i t y i s t o be o p t i mized. Commonly an o r g a n i z a t i o n such as a pharmaceutical company has c e r t a i n major areas of i n t e r e s t (e.g., a n t i m i c r o b i a l s , cont r a c e p t i v e s , e t c . ) w i t h i n which new lead s t r u c t u r e types are sought. For each such area of i n t e r e s t , a set of b i o l o g i c a l t e s t s o r screens i s designed s p e c i f i c a l l y t o t e s t f o r the d e s i r e d a c t i v i t y . The lead compounds are normally found e i t h e r through random t e s t i n g i n these screens of d i v e r s e chemical types or through ideas t r i g g e r e d by p u b l i c or in-house l i t e r a t u r e . The s e l e c t e d lead i s then developed by a c y c l i c process i n which analogs are s y n t h e s i z e d and bioassayed, r e s u l t s are anal y z e d , and new analogs p o s t u l a t e d t o have even g r e a t e r a c t i v i t y are proposed f o r s y n t h e s i s . This i s the t r a d i t i o n a l realm of m e d i c i n a l chemistry and the area i n which most of the work on m e d i c i n a l chemical i n f o r m a t i o n r e t r i e v a l has been done. Novel compounds s y n t h e s i z e d i n the analog e v a l u a t i o n may be patented a t any p o i n t , and f o r p a r t i c u l a r l y promising compounds, a d d i t i o n a l analogs may be prepared f o r patent p r o t e c t i o n . Compounds t h a t show p a r t i c u l a r l y good a c t i v i t y i n the p r i mary b i o l o g i c a l screening t e s t s are submitted t o pathology, t o x i cology, and pharmacology s t u d i e s to f u r t h e r d e f i n e t h e i r s u i t a b i l i t y as drugs. For those that show promise f o r c l i n i c a l use app r o p r i a t e f o r m u l a t i o n s are developed. When a compound s a t i s f a c t o r y i n a l l of these areas has been found, a request ( I n v e s t i g a t i o n New Drug a p p l i c a t i o n , o r IND) t o t e s t the drug i n humans i s submitted t o FDA summarizing a l l e x i s t i n g data on the compound. More or l e s s c o n c u r r e n t l y , process development s t u d i e s are undertaken t o optimize the manufacturing process and f o r patent prot e c t i o n . Marketing i m p l i c a t i o n s may a l s o be considered i n further d e t a i l at this point. When the IND a p p l i c a t i o n i s approved, c l i n i c a l t r i a l s can be



Figure 1.

LEAD SELECTION oMASS SCREENING oLITERATURE

MANAGEMENT

MARKETING

MANUFACTURING r

*

CLINICAL TRIALS

FDA APPLICATION (IND)

FDA APPLICATION (NDA)

PROCESS DEVELOPMENT

FORMULATION

PHARMACOLOGY

PATHOLOGY/ TOXICOLOGY

Major functions in a typical drug-development program

POST MARKETING MONITORING

PATENT

BIOASSAY (LEAD DEVELOPMENT)

REGULATORY A F F A I R S

EVALUATION/ PREDICTION

ANALOG SYNTHESIS AND CHEMICAL ANALYSIS

-I, INVENTORY h —

PUBLICATION

4


i n i t i a t e d . Concurrently, manufacturing f a c i l i t i e s and a marketing program are e s t a b l i s h e d . I f the c l i n i c a l t e s t i n g i s successf u l , the r e s u l t s are submitted to the FDA along w i t h a New Drug A p p l i c a t i o n (NDA) requesting approval of the drug f o r use i n the general p o p u l a t i o n . F i n a l l y , once the product i s on the market, i t s use i s monitored to detect any a d d i t i o n a l i n d i c a t i o n s of use or adverse e f f e c t s t h a t may not have been evident during o r i g i n a l testing. In a d d i t i o n to these d i r e c t e d f u n c t i o n s , three types of f u n c t i o n s e x i s t which pervade e s s e n t i a l l y a l l of the other funct i o n s of Figure 1: (1) the management f u n c t i o n i n which d e c i s i o n s are made to determine which p r o j e c t s or compounds w i l l progress, which w i l l be delayed or terminated; (2) the p u b l i c a t i o n f u n c t i o n i n which r e s u l t s are provided to the i n t e r n a l and open l i t e r a t u r e ; and (3) the f u n c t i o n of responding to government regulations r e l a t i n g t Information Needs Related to the Drug Development Process Development of a safe and u s e f u l drug, which i s the u l t i m a t e goal of m e d i c i n a l chemistry, i s an extremely complex and c o s t l y process. The purpose of m e d i c i n a l chemical i n f o r m a t i o n r e t r i e v a l i s to support t h i s process, to permit r a p i d and accurate i d e n t i f i c a t i o n of c l i n i c a l l y u s e f u l compounds w i t h minimal r i s k , c o s t , or delay. The types of i n f o r m a t i o n r e q u i r e d to provide t h i s support are extensive and d i v e r s e . The exact types of data needed w i l l be discussed i n more d e t a i l l a t e r , but some g e n e r a l i z a t i o n s can be made about the way the data must be used. The b a s i c data f u n c t i o n s r e q u i r e d are i n f o r m a t i o n storage, r e t r i e v a l , a n a l y s i s , and r e p o r t i n g . The o r g a n i z a t i o n of the data i n t o v a r i o u s f i l e s must be balanced f o r maximum e f f i c i e n c y . A l though each of the f u n c t i o n s of Figure 1 has i t s own primary i n t e r e s t i n f o r m a t i o n , many s i t u a t i o n s r e q u i r e the combined use of data from d i f f e r e n t areas. For example, the management, p u b l i s h i n g , and r e g u l a t o r y a f f a i r s f u n c t i o n s r e q u i r e access to n e a r l y a l l of the data types at one p o i n t or another. Thus, w h i l e the data must be segmented to a l l o w e f f i c i e n t access by i t s prime users, i n t e r f a c e s must be provided to s a t i s f y c r o s s - f u n c t i o n a l needs as w e l l . The q u a l i t y of the data i s a v i t a l c h a r a c t e r i s t i c . " Q u a l i t y " here encompasses not only accuracy, but comprehensiveness ( i n c l u s i o n of o l d and new data) and s u i t a b i l i t y ( p r o v i d i n g the exact type of data needed r a t h e r than some nebulous f u n c t i o n or q u a l i t a t i v e estimate t h e r e o f ) . Comprehensiveness i s p a r t i c u l a r l y important where chemicals w i t h human b i o a c t i v i t y are being prepared. Procedures f o r using the system should be as a t t r a c t i v e and as simple as p o s s i b l e to encourage d i r e c t use by s p e c i a l i s t s w i t h i n each f u n c t i o n who may not a l s o be i n f o r m a t i o n s p e c i a l i s t s . I n t e r a c t i v e o p e r a t i o n , simple commands, f l e x i b l e outputs t h a t are


1.

MILNE ET AL.

Medicinal

Chemical

Information

5

f a m i l i a r t o the user (such as standard t e x t r a t h e r than computer codes, standard s t r u c t u r a l diagrams r a t h e r than l i n e a r n o t a t i o n s ) a l l help t o get the i n f o r m a t i o n d i r e c t l y i n t o the hands of the i n d i v i d u a l s best equipped t o use i t . Although the p r o v i s i o n of i n t e r f a c e s s u i t a b l e f o r non-computer s p e c i a l i s t s adds c o n s i d e r a b l y to system development and o p e r a t i o n a l c o s t s , these costs should be f a r outweighed by the r e s u l t i n g i n c r e a s e i n e f f e c t i v e n e s s of the t o t a l program. (Of course, some f u n c t i o n s are s t i l l unavoidably complex and w i l l r e q u i r e i n t e r m e d i a t i o n by an i n f o r m a t i o n s p e c i a l i s t f o r the f o r s e e a b l e f u t u r e . ) O r g a n i z a t i o n and Current Status of M e d i c i n a l Chemical Information The major data types r e q u i r e d f o r m e d i c i n a l chemistry as p a r t of a t o t a l drug development program are i n d i c a t e d i n Figure 2. I n this conceptualizatio around the chemical compound ber ( g e n e r a l l y an i n t e r n a l r e g i s t r y number). Four c a t e g o r i e s of data are d e f i n e d : chemical data and b i o l o g i c a l data, which chara c t e r i z e the compound i t s e l f , management/distribution data which c h a r a c t e r i z e the commercial aspects of the compound, and secondary or b i b l i o g r a p h i c data which i n essence are p o i n t e r s t o chemical, b i o l o g i c a l , or management/distribution data i n the open (or occasionally, internal) literature. (a) Chemical Data. V i r t u a l l y a l l o r g a n i z a t i o n s t h a t support a major drug development e f f o r t have a computerized f i l e of the chemical s t r u c t u r e s considered i n the program. F i l e s on the order of 100,000 - 400,000 s t r u c t u r e s are not uncommon. The s t r u c t u r e s are represented as e i t h e r connection t a b l e s , l i n e n o t a t i o n s , o r a t t r i b u t e codes. At present the f i l e s are used f o r s e v e r a l purposes, the most common of which are (a) d u p l i c a t e checking t o determine i f a compound has already been t e s t e d , (b) s u b s t r u c t u r e searching to s e l e c t i v e l y r e t r i e v e compound c l a s s e s , and (c) d i s play of two dimensional s t r u c t u r a l diagrams. Of the chemical data c a t e g o r i e s i n F i g u r e 2 the molecular s t r u c t u r e r e p r e s e n t a t i o n has r e c e i v e d by f a r the g r e a t e s t emphasis i n c u r r e n t i n f o r m a t i o n systems. The other data types ( a n a l y t i c a l , physicochemical, process development, s y n t h e s i s / r e a c t i v i t y ) have been d e a l t w i t h i n d i v i d u a l l y , but are only beginning to be i n c o r porated w i t h the s t r u c t u r e f i l e s t o form i n t e g r a t e d systems. (b) B i o l o g i c a l Data. V i r t u a l l y a l l major m e d i c i n a l chemical i n f o r m a t i o n systems have automated f i l e s which c o n t a i n the primary t e s t i n g r e s u l t s . Because of the importance of c o r r e l a t i n g and coo r d i n a t i n g b i o l o g i c a l data w i t h s t r u c t u r a l data i n the analog development c y c l e , automated l i n k s between the s t r u c t u r e data and primary t e s t i n g data have been provided i n some systems. However, the sheer volume of the b i o l o g i c a l t e s t r e s u l t s (and i n some cases, a d m i n i s t r a t i v e c o n s i d e r a t i o n s ) has c o n t r i b u t e d t o the slow



Figure 2. The medicinal chemical information network. The types of data required in a total drug-development process can be viewed as organized around the compounds tested, with four associated categories of data: chemical, biological, management/distribution, and secondary.

1.

MILNE ET AL.

Medicinal

Chemical

Information

7

progress i n t h i s area. I n many systems the i n t e r f a c e between the two data f i l e s i s e i t h e r u n w i e l d l y o r n o n e x i s t e n t . As was the case w i t h chemical data, the remaining forms of b i o l o g i c a l i n f o r m a t i o n ( t o x i c o l o g y , pathology, pharmacology, c l i n i c a l , f o r m u l a t i o n ) have a l l undergone a c e r t a i n degree of automation, but the extent t o which these data are i n t e g r a t e d w i t h chemical s t r u c t u r e s and primary screening data v a r i e s cons i d e r a b l y . As noted above, a major c o n s t r a i n t appears t o be the volume of data t h a t must be encompassed by these b i o l o g i c a l s y s tems. I n t e r f a c i n g them w i t h each other and w i t h chemical i n f o r mation f o r purposes of r e t r i e v a l can place a heavy burden on even very l a r g e computer systems. As a r e s u l t , users must commonly r e l y on manual methods f o r c o o r d i n a t i n g the data from d i f f e r e n t areas. (c) Management/Distribution p o r t i o n of the m e d i c i n a c a t e g o r i e s of data which a r e extremely important t o the o p e r a t i o n of an e f f e c t i v e drug development program. The d i f f e r e n c e s between e x i s t i n g systems are g r e a t e s t i n t h i s area, i n terms of the data types t h a t are handled and the c a p a b i l i t i e s and i n t e r f a c e s to the r e s t of the network that are provided. Some o f these data types (e.g., i n v e n t o r y and compound sources) are g e n e r a l l y automated and w e l l - i n t e g r a t e d w i t h the chemical and b i o l o g i c a l data. Patent i n f o r m a t i o n on in-house compounds may be automated but i s not commonly l i n k e d d i r e c t l y to the chemical and b i o l o g i c a l data. Manufacturing i s u s u a l l y a separate i n f o r m a t i o n category, not l i n k e d t o the r e s t of the system. Except i n s p e c i a l cases, such as manufacturing problems r e l a t e d t o the chemistry of a p a r t i c u l a r process, t h i s s e p a r a t i o n i s reasonable a t present. Marketing i s another area which c u r r e n t l y appears t o be d i vorced from the r e s t of the network. The importance of market a n a l y s i s (both before and a f t e r approval of a product) as an i n t e g r a l p a r t of the drug development e f f o r t i s becoming i n c r e a s i n g l y apparent. Such a n a l y s i s i s v i t a l d u r i n g lead s e l e c t i o n t o determine what products are needed, during process development t o guarantee reasonable manufacturing c o s t s (where "reasonable" i s r e l a t e d t o the drug's market), and d u r i n g the i n i t i a l stages of p u b l i c use t o respond t o any e f f e c t s t h a t may not have been e v i dent during t e s t i n g . One of the most a c t i v e areas i n the handling o f m e d i c i n a l chemical i n f o r m a t i o n i s concerned w i t h the requirements of government r e g u l a t o r y agencies. This i n c l u d e s i n f o r m a t i o n r e quired not only f o r purposes o f drug approval but a l s o t o comply w i t h r e g u l a t i o n s concerning the environmental e f f e c t s of the manu f a c t u r i n g process and the h e a l t h and s a f e t y of i n d i v i d u a l s exposed t o chemicals i n any phase of the drug development process. In many o r g a n i z a t i o n s major e f f o r t s a r e now being undertaken t o i n c o r p o r a t e h e a l t h , s a f e t y , and environmental data i n t o the t o t a l chemical i n f o r m a t i o n network.


8


(d) Secondary Data, Almost a l l drug development o r g a n i z a t i o n s u t i l i z e one or more of the major o n l i n e i n f o r m a t i o n s e r v i c e s f o r r e t r o s p e c t i v e and current awareness l i t e r a t u r e review. I n add i t i o n , in-house i n f o r m a t i o n groups are o f t e n r e s p o n s i b l e f o r r e viewing the current l i t e r a t u r e on s p e c i f i c problems. Most c a t e g o r i e s of data that are c u r r e n t l y r e t r i e v a b l e through the o n l i n e s e r v i c e s are t e x t o r i e n t e d (such as a b s t r a c t s , keywords, and patent d a t a ) , however i n c r e a s i n g l y more s p e c i a l i z e d data types are being o f f e r e d , such as p h y s i c a l constants, r e a c t i o n s , b i o l o g i c a l data (e.g., LD50), and substructure i n f o r m a t i o n . Some i n t e g r a t i o n of these f i l e s e x i s t s , though a t present the i n t e g r a t i o n i s p r i m a r i l y w i t h i n a s i n g l e i n f o r m a t i o n assembling o r g a n i z a t i o n . Nevertheless, there appears to be a growing r e c o g n i t i o n of the need f o r i n t e r f a c i n g and c o m p a t i b i l i t y even among competing s e r v ices . Government supporte of p u b l i c i n f o r m a t i o n . respond to government r e g u l a t i o n s as w e l l as to a s s i s t t e c h n i c a l researchers. Data bases c o n t a i n i n g s t r u c t u r e s of p e r t i n e n t compounds (e.g., r e g u l a t e d drugs, carcinogens, t o x i c compounds) are r a p i d l y being b u i l t , and sometimes i n c o r p o r a t e a d d i t i o n a l u s e f u l chemical i n f o r m a t i o n such as c r y s t a l l o g r a p h i c and s p e c t r a l data. An i n t e r e s t i n g s i d e e f f e c t to the development of these f i l e s has been the encouragement of c o m p a t i b i l i t y among p r i v a t e systems. This has r e s u l t e d because a number of p r i v a t e o r g a n i z a t i o n s are seeking uniform methods of accessing in-house and p u b l i c f i l e s and have standardized on the format of the p u b l i c systems. Future Trends i n M e d i c i n a l Chemical Information While the f o l l o w i n g chapters i n t h i s book present a comprehensive view of the current c a p a b i l i t i e s of m e d i c i n a l chemical i n formation systems, they a l s o provide i n s i g h t s i n t o the d i r e c t i o n s of progress of the f i e l d as a whole. The foremost trend i s toward i n t e g r a t i o n : the p u l l i n g t o gether of d i s c r e t e in-house systems and the c r e a t i o n of automated i n t e r f a c e s t o p u b l i c and government systems along the l i n e s of the i n f o r m a t i o n network i n Figure 2. Much of the stimulus f o r i n t e g r a t i o n comes from the c r o s s - d i s c i p l i n a r y nature of the t e c h n i c a l i n f o r m a t i o n needs (such as the need f o r simultaneous access to chemical and b i o l o g i c a l data i n the analog development p r o c e s s ) . At the same time, as the i n f o r m a t i o n systems become more encompassing and more responsive to the t e c h n i c a l needs, they are ena b l i n g better-informed and more coordinated management d e c i s i o n s at higher l e v e l s . Such support of the management f u n c t i o n i s expected g r a d u a l l y to become more formal, o f f e r i n g management s p e c i f i c r e p o r t i n g and i n q u i r y c a p a b i l i t i e s w i t h simultaneous access t o a l l types of data on demand. Another extremely important trend i s toward greater end-user o r i e n t a t i o n . This area was somewhat neglected w h i l e system de-


1.

MILNE ET AL.

Medicinal

Chemical

Information

9

velopment e f f o r t s focused on techniques f o r data storage and r e t r i e v a l . The emphasis on end-users i s evident i n the increased a p p l i c a t i o n of computer graphics t o the user-computer i n t e r f a c e and i n the growing number of i n t e r a c t i v e systems. Greater f l e x i b i l i t y i n adapting t o s p e c i a l i z e d needs i s apparent not only i n the v a r i e t y of data types and data manipulations t h a t are poss i b l e , but a l s o i n the o r g a n i z a t i o n and r e o r g a n i z a t i o n of outputs to permit data t o be examined from many s t a n d p o i n t s . The r e t r i e v e d data i s more end-user o r i e n t e d i n order t o be interprétable without cumbersome lookup t a b l e s o r other a r t i f a c t s of comp u t e r i z a t i o n . O v e r a l l , the apparent o b j e c t i v e i s to make i n f o r mation systems a working t o o l of the i n d i v i d u a l s best equipped t o use them, r a t h e r than a reference u t i l i t y a c c e s s i b l e only through the i n f o r m a t i o n s p e c i a l i s t . Now t h a t automated r e t r i e v a l i s w e l l e s t a b l i s h e d , e f f o r t s are f o c u s i n g on automate r a t e s t a t i s t i c a l and h e u r i s t i cated f u n c t i o n s such as quantum mechanics and conformational anal y s i s are being i n t e r f a c e d d i r e c t l y t o l a r g e f i l e s . I n t e r e s t i n computer-aided s y n t h e t i c a n a l y s i s , computer-aided s t r u c t u r e e l u c i d a t i o n , and computer-aided process development remains h i g h , but as these complex a p p l i c a t i o n s are s t i l l i n the a c t i v e development stages they have not yet undergone f u l l - s c a l e i n t e g r a t i o n i n t o e x i s t i n g i n f o r m a t i o n networks. Summary This chapter has o u t l i n e d the nature of the drug development process and has described m e d i c i n a l chemical i n f o r m a t i o n r e t r i e v a l i n terms of a network of d r u g - r e l a t e d i n f o r m a t i o n c a t e g o r i e s . A l l of the systems described i n t h i s volume can be viewed i n terms of t h i s o r g a n i z a t i o n . Some of the systems, notably those of the major pharmaceutical companies and some of the l a r g e r government programs, encompass a l a r g e f r a c t i o n of the drug development funct i o n s and data types. Other systems may i n c l u d e only a s i n g l e f u n c t i o n and r e l a t i v e l y few data types, yet even these s m a l l e r (or l e s s i n t e g r a t e d ) systems feed i n t o and are p a r t of the t o t a l med i c i n a l chemical i n f o r m a t i o n network. This i n t r o d u c t o r y chapter has provided only an overview of the f i e l d and a framework f o r viewing the s p e c i f i c c a p a b i l i t i e s that e x i s t . The d e t a i l s and exact mechanisms are provided i n the papers t h a t f o l l o w and i n the references c i t e d t h e r e i n . RECEIVED August 29, 1978.


2

The

PRODBIB

D a t a Base: R e t r i e v a l of P r o d u c t

I n f o r m a t i o n f r o m the P u b l i s h e d Literature

BARBARA C. FREEDMAN Product Information Section, Technical Information Department, Burroughs Wellcome Co., 3030 Cornwallis Road, Research Triangle Park, NC 27709 The L i b r a r y at Burrough charged w i t h the responsibility retrieving published scientific and medical literature (hereafter r e f e r r e d to as "product papers") about Company products. The references c o l l e c t e d have been used in two ways: (1) to produce b i b l i o g r a p h i e s to accompany I n v e s t i g a t i o n a l New Drug a p p l i c a t i o n s , New Drug A p p l i c a t i o n s , and Annual Reports to the Food and Drug A d m i n i s t r a t i o n ; and (2) to support the Research, Development, M e d i c a l , and Marketing D i v i s i o n s in their ongoing work w i t h Company products. Until 1972 this responsibility was discharged using entirely manual methods. Three files e x i s t e d : (1) a card file, organized by product and, w i t h i n product, by author; (2) copies of b i b l i o g r a p h i e s prepared from these cards s i n c e 1966; and (3) r e p r i n t s of articles published s i n c e 1970. About 25,000 references had been c o l l e c t e d and the files were growing at the r a t e of approximately 5000 per y e a r . Searching these manual files became i n c r e a s i n g l y difficult, as d i d the production of annual b i b l i o g r a p h i e s . In 1972 we s e l e c t e d INQUIRE (Infodata Systems Inc., Falls Church, Virginia) as an appropriate software package to handle a data base of our own d e s i g n , i n t o which we could enter and retrieve b i b l i o g r a p h i c i n f o r m a t i o n , subject i n d e x i n g , and data e x t r a c t e d from the product papers. We c a l l the data base PRODBIB, f o r "product b i b l i o g r a p h y . " Our s e l e c t i o n of INQUIRE was based on the s u c c e s s f u l experience o f other pharmaceutical firms (1). Hardware. From November 1972 to November 1975 we ran our INQUIRE data bases at T r i a n g l e U n i v e r s i t i e s Computation Center, a l a r g e , university-owned computation c e n t e r . In November 1975 we moved INQUIRE operations to our own Computer S e r v i c e s D i v i s i o n , where we have an IBM 370/148, and run INQUIRE under 0S/VS1, i n batch mode. 0-8412-0465-9/78/47-084-010$05.00 © 1978 American Chemical Society In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

2.

FREEDMAN

PRODBIB

Data

Base

11

INQUIRE f a c i l i t i e s . We are c u r r e n t l y using V e r s i o n 9.1, w i t h blocked f i l e s and the m u l t i - d a t a base o p t i o n . O r g a n i z a t i o n of the PRODBIB f i l e . An item i n the PRODBIB f i l e c o n s i s t s o f the b i b l i o g r a p h i c data and indexing terms f o r one product paper. I f two o r more products are mentioned i n one paper, however, we create one item f o r each product. I d e n t i c a l b i b l i o g r a p h i c i n f o r m a t i o n i s entered f o r each, w i t h indexing terms appropriate f o r each product. Items are s t r u c t u r e d according to the f i e l d s d e f i n i t i o n t a b l e (Figure 1). The only r e q u i r e d f i e l d i n the data base i s REPORT, a r e p o r t code and accession number assigned to each paper. For papers which mention more than one product, a DOCPT (a s i n g l e a l p h a b e t i c c h a r a c t e r ) i s assigned. Consequently, REPORT uniquely i d e n t i f i e s the product paper; REPORT and DOCPT uniquely i d e n t i f y a PRODBIB item The f i e l d s PATIEN graphic i n f o r m a t i o n , a quick c a t e g o r i z a t i o n of each paper, and the number of p a t i e n t s i n v o l v e d i n each study. The DRUG f i e l d contains the N a t i o n a l Drug Code number, B.W. Co. compound number, and B.W. Co. tradename f o r the product being indexed. We use the tradename i n the DRUG f i e l d only and generic names i n the remaining f i e l d s . I n t h i s way we can d i s t i n g u i s h i n searching between the product as the subject o f an item and the product used together w i t h or compared to another product. The remaining f i e l d s are used f o r subject i n d e x i n g . For almost a l l of these we use terms s e l e c t e d from MALIMET, the thesaurus of the Excerpta Medica Foundation (2). MALIMET i s a broad and deep l i s t of terms, w i t h f a i r l y good c o n t r o l and a c e r t a i n amount o f s t r u c t u r e . ( I t i s not, however, h i e r a r c h i c a l . ) We have used about 8000 unique terms from MALIMET. Figure 2 shows the content of a t y p i c a l PRODBIB item. Indexing p o l i c y . I t i s our p o l i c y to index only that i n f o r m a t i o n i n a paper r e l e v a n t to our product. We do not index the whole content of a paper, as MEDLARS would, but r a t h e r the narrower range of i n f o r m a t i o n about our product. The primary advantage of t h i s approach i s that we can e l i m i n a t e many f a l s e drops that occur whem we search the " g l o b a l " data bases. Keywords are posted against f i e l d names i n such a way as to create a keyword-in-context s i t u a t i o n , i . e . , a keyword can l a t e r be searched i n terms of the f i e l d i n which i t was entered. For example, the keyword RASH can be used as an i n d i c a t i o n f o r one product, an adverse r e a c t i o n to another product, and a c o n t r a i n d i c a t i o n f o r a t h i r d product. In s e a r c h i n g , the keyword can be used alone or as a f i e l d value c o n d i t i o n of a p a r t i c u l a r f i e l d .



12 OISPLAY

F I tLDS.

FIELD NAME

KEY

TYPE

STORED LENGTH

STRUC

- PRINT FORM LEN

RPTS

NOTES

UAIAuASc « PROJBIB · REPORT SJUkCE UUCYR JOCNU DUCPT PAT1ENI S Pub γ κ INPUT LONDON INDEXER DRUo OKUGNU JRUGLBLk DRUGPROD JRUGPKG CuMPUuNJ CMP i) NO CMPObRC CMPYK CMPBATCH PRODNAML AUTHùk

PFX PFX PFX

PFX

PFX

PFX

CHR LHR ChR CHR CHR INT CHR i NT CHR CHR CHR CHR CHR CHR CHR CHR

12 4

2 4

1 4

2 6

6 3 38 9

3 * 2 9

CHR

CHR CHR

1 1TL6

CHR

V

CITATΙυΝ L ANGUAGL LULAIIUN CAT E G U R Y TYPE ΙΝυΙοΔΤΝ EFFLC1 LGNCuM CUMPAKE cxpsutij EXPTYP 1 EXP Τ Y P 2 ΕΧΡΝι, tXPStlX I N V 1 Iku PRtPN PREPSKL FORMULA FUkM kAT I C OkUoL CONCl 0RUG2 CUiMC^ DOSAbE DUSfc ROUTE ADVERSE ADVREAC ADVNUM CUNTRINU SENSURG RESISUKG SUBJECTS EFF ICACY NOTES KEYS ITEMNU SFIXEO ALLKEYS

CHR CHR

V

CHR PFX PFX

CHR CHR

SMP

CHR

SVP SMP

LHK

SMP

CHR

2 2 20 24 280 168 15 60 12 9

CHR

3 71 P R E S E N T P k O b L t h S OF G u U T . C L I N I C A L b I CCHc M I :> I RY THfcRAPI EwCChE 2 2 ( 2 ) : 77-8*, 1972 TRANSLATION GERMAN GUNTHtk

AU LNUL i POMATOS IS)

k

UF

oJUl

lL:W7o/3l94

Ε

CLINICAL

FINDINGS

ME T A B O L I C

EFFECTS

AND THERAPY OF

OF

oOUT

SULFINPYRAZONE

WITH S P E C I A L

(ANTUkAN)

REFERENCE

10

THE

AND A L L OPURINOL

(ZYL0R1C) WlEN

K L I N WOCHENSCHR GERMAN

Figure 8.

PRODBIB

Hi:

817-320

NOV 7 ,

1969

search using field value conditions (search specification followed by search results)


2.

FREEDMAN

PRODBIB

Data

CLINICAL

Base

USES

17

OF A L K E R A N

INOICATIUN

ITEMS

OSTEOSARCOMA OVARY A O E N O C A R C I ΝCMA OVARY CANCER OVARY CANCER M E T A S T A S I S OVARY CARCINOMA OVARY C Y S T A u c N U C A R C I NOMA OVARY TERATOMA OVARY TUMOR PANCREAS AUENOCARCINGMA PANCREAS CANCER

CLINICAL

25 2 36 I 38 2 1 2 I I

USES

OF A L K E R A N

M U L T I P L E MYELOMA L A T O S OL V A L E N T I N E AM TREATMENT

VIRGINIA

OF

JLÙ0/73/0369 HYPEKCALCE'-I

M F13

J 69(3):

1A * i i h

52-6*

F'.JROStMloh

MARCH

AIML>

COk I I C u o I L R O I O S

197;

M U L T I P L E MYELuMA LAW IP TLbW 7θ/ϋο09 PLOVNICK H J ùEOUOw OG MOL Τ I PL Ε M Y E L O M A , S I J i _ R u b L A S I IC Α Ν Γ . Μ Ι Α Α Ν ύ Α ν , υ Τ Ε Ν E N u L J MEO 2 9 * ( 3 ) : l o ^ JA.N 1 5 , 1 9 / 6 L E I Τ Εκ

LEUKEMIA

M U L T I P L E MYELOMA LAw M t P T L < W 7o/106i> F A M I L I A L OCCURRENCE OF M U L T I P L E MYcLOMA SOUTH M c J J 6 9 ( 1 ) : 4 6 - t . J J A N 1 9 7 6

M U L T I P L E MYELOMA LE C H E V A L L l E k PL T k t Α ί MENT OF M Y t L L M A . SEM HOP P A k i S 4 / : 2 * 9 - 2 3 1 FRENCH

Figure 9.

PRODBIB

TL8U//2/G226 J A N 20,1971.

search listing réfrénées

wtih a table of contents



18 ADVERSE

REACTIONS

TO S E P T R A

FOLIC ACID DEFICIENCY BUSHbY SRM I f< I M E T h ù P r I Μ A N D SUE P H U I N A M I D E S : S

FOLIC

At- R

MED J

iVCI'J

CLARK

44(SUPPL):

3 - 1 0 AUu

TLbD/74/103ô LAoJRATJRY STJOHS 15,

197J

DEFICIENCY F

TL3D/76/4108

i M u G S MI\J V I T A M I N DEFICIENCY A O V E K S L DkJO FFAC1 bULL 5 7 : 1 9 o - 1 9 9 ,

FOLIC

ACI J

C L L MAN HERBERT

JLFICIENCY Ν

TL30/76/419 7

V

C u I k I MU X A Z J L Ε

LANLcï

FOLIC ACIu DAVIS JALKSLN

2« Œ>

»

co

CO

5

™

4»

Ο

u

υ

; ^

'-'

+

2

Ξ ~

I

M

"

a S

ο +

> X

^

^.

2 S * <s

υ .,

.,

S)

S3

χ "* _

~

'£

i r

m ο

s S o s

· §

2"'

u

ω ω ο χ χ

*L S3

s,

s,

Si

S3

S3

(β a =

rr. CM CM VO

Ο

£

D C Ù U O X

-

-

* „ ^ * ic s i c ? sis

S

s

S

3

ir

^ . ^

^t £ s ^ 1

tt

υ

w

'-'

S

5

s * ^ <s s

**

rr

ρο

+

5

0

<s

« H*. ς» S~>

^

«2

ω

:

ww υ M

* £ α»

^

s

UU3CQ

v u ^ ai a Ξ

^ νο

S

S

CO

ΰ

CS

2

^

^

!ί? «Ni

Γ

Ο w. s-^

Ό

X

2

+

* «S,

«M

CM

w

S SSSΚSSSS(SS ζ$ C"2 Κ s s s s g VCSJVOK)^ r- r -8

u ο

vO

+

4 U >. <s + ω ® 1 ο

_ Ο

**•*

>—' ι—' r-" (NI CM

* <S

Ο

«S

s S

f-> f t— (NI rr r r s s κ si s s

n

M

ο,

U

s w

CC 0Γ 0. CO CLl ce

, HOURS 25,3184 β,7457 5.3350 11.2079 CONTROL 59,3786 21.7706 7.2409 25.0231 ORUG 1.9059 13.8152 13,0249 DELTA 34,0602 10.0319 2.9156 10,0495 11,6807 T»CALC PROB 0,0000 0,0000 0.0000 0,0106 0*24 HOURS 10.2726 20,2672 49,1386 19,7823 CONTROL 62.1130 29,8380 12.3537 31.9110 DRUG DELTA 32,9744 10,0557 11,6438 2.0*11 9.7624 8,1002 2,337? 6.1872 T-CALC 0.034* 0,0000 PROB 0,0000 0,0000 2·6 HOURS 2,7625 6.5274 5.1353 CONTROL 14,3502 13.0482 30,3568 11,7863 3,5197 ORUG 0.7*74 6,5208 6.6510 DELTA 16.0066 9.4229 2.7*17 7.7320 9.5962 T»CALC PROB 0,0000 0,0000 0.0144 0,0000 §•24 HOURS CONTROL DRUG DELTA T-CALC MOB

23,8203 22,7345 •1,0858 •0,3532 0,7292

11.0365 β,0674 -2,9691 •2,2714 0,0394

4.9376 5.Π2Λ 0.1752 0,2Λ2« 0.7BH

9.0592 6,8879 -2.1713 •1,9738 0.0685

PAGE

NA/K RATIO

EXPT

URIC AC 10 (MG/KG)

#00378

6,2469 0.0000

-1.3190 0.2083

1.6776 3.0149 1.3373 9.5873 0,0000

7.0379 5.9875 •1.0504 •3.4965 0,0036

2,0004 2,4113 0,4109 2.4598 0,0275

26.1265 21.0431 •5.0834 -3.5757 0.0030

1,8887 3.3591 1.4705 10.5562 0.0000

4.1634 3.4475 -0.7159 •2.2744 0.0392

2.3939 1.5658 -0.8281 -2,4906 0.0259

19.0886 15.0556 •4.0330 •2.8200 0,0136

KITH THE EXCEPTIONS, \f ANY, NOTED BELOW FXPERIHENTS §00368 · #00381 ARE OK FOR ENTRY INTO ABCO I I

Figure 8.

, -, 8I6NE0

9

— — DATE

Sample statistical report



38

Summary The two approaches described in this paper for automation of test data transmission from the source to a computer master file have decreased the cost of data entry and increased the accuracy of data recording. With a totally computerized operation, changes in experimental parameters and/or statistical calculations on raw data can be accomodated by modification of the programs. This allows the scientist flexibility in recording results of revised experimental procedures. Abstract In order to eliminate, wherever possible, keypunching of large volumes of data per experiment or transcription of results obtained from computer statistical calculations, the experimenters enter origina computer terminal. A l controlled by computer programs; humans intervene only when required for visual evaluation of the information or approval of outputted results. This approach is cost effective and ensures producer responsibility for recorded information. Two methods are discussed: 1. Direct recording of observations (e.g. plant growth) on a portable battery powered recording device, followed by data transmission to and further processing on the R&D computer; and 2. CRT entry of observed values for test and control animals for specific tests, followed by appropriate statistical calculations with report generation and simultaneous reformatting of summary information for the corporate master data base (ABCD). Only after verification of the accuracy of the data by the producers of the information are the results entered into the master data base. Acknowledgements The authors wish to thank the following for their assistance and cooperation in these projects: Dr. Amrit L a l l , Patricia Morse, Karen Oheim, Douglas Reno, and Donald Weber. Literature Cited 1. J.

Morphis, Β. B , Torbet, Ν . , Hunter, W. W., and Broome, F . K., Chem. Doc. (1966) 6, 77-81.

2. Ginsberg, H. F., G r e t h , P . Α . , and Morphis, B . B.,"ABCD II A User C o n t r o l l e d Biological - Chemical Data System," Presented before the Division of Chemical Literature, ACS, New York, N.Y., August 28, 1972. 3.

Federal R e g i s t e r , November 19, 1976, 41 (225), 51206-51229.

4. MSI Data C o r p o r a t i o n , 340 F i s c h e r 92627

Avenue,

Costa

Mesa, CA

RECEIVED August 29, 1978.


4

C l u s t e r i n g in F r e e - T e x t D a t a Bases

RUDOLPH J. MARCUS, EDWIN T. FLORANCE, and EUGENE E. GLOYE Office of Naval Research, 1030 East Green Street, Pasadena,CA91106

In previous paper th author hav demonstrated th utility of t e x t h a n d l i n g method structural i n f o r m a t i o n . Although the methods developed have general applicability, the particular data base used was one composed of medical and chemical information and led to heuristic structure-activity correlations. Specifically, the data base c o n s i s t s of all of the e n t r i e s from the eighth e d i t i o n of the Merck Index which list a medical use. There are 3,433 such compounds. For each of them we have listed all of the medical uses as well as all of the synonyms by which these compounds are known. In a d d i t i o n t o trade names, these synonyms c o n t a i n one or more Geneva system names. These Geneva system names c o n t a i n the "structural" i n f o r m a t i o n . (The Geneva system of nomenclature has been u n i v e r s a l l y used by chemists s i n c e 1864 and is kept up-to-date as new nomenclatural exigencies arise. In the Geneva system, each syllable u n i q u e l y defines a structural module of a m o l e c u l e , and the position of the syllable in the name tells how the modules fit together i n the m o l e c u l e . ) I t i s the m a n i p u l a t i o n of v a r i o u s Geneva system names as t e x t which c o n s t i t u t e s part of the n o v e l t y of our work. M a t e r i a l q u a l i f y i n g the medical use terms such as a former use or an experimental use has been coded i n t o the data base. Intere s t i n g c o n c l u s i o n s from the d i s t r i b u t i o n of such use q u a l i f i e r s w i l l be l i s t e d below. I t i s seen, then, that our data base i s a c l o s e d , s e l f - c o n s i s t e n t u n i v e r s e which i s not updated. Simple r e t r i e v a l i s p o s s i b l e by searching e i t h e r by s t r u c t u r a l module or by medical u s e . Exhaustive study of the medical use part of the data base began w i t h the counting of uses and the concurrent c o m p i l a t i o n of an i n v e r t e d Merck Index which could be entered by medical use r a t h e r than by chemical compound. In that manner a " s o c i o l o g y " of medical use language was derived (1). Because both medical use and chemical s t r u c t u r e are assoc i a t e d i n the same computer f i l e , they form a hyperspace. The nonparametric nature of t h i s hyperspace was discussed i n a This chapter not subject to U.S. copyright. Published 1978 American Chemical Society In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

RETRIEVAL

40

O F MEDICINAL

CHEMICAL

INFORMATION

previous paper (2) and a r i s e s from the n a t u r a l language t e x t , r a t h e r than n u m e r i c a l , nature (alphanumeric r a t h e r than numeric) of the data base. C l u s t e r s i n such t e x t systems were defined i n that paper as a c o l l e c t i o n of v e c t o r s whose column elements are s i m i l a r . Mutual e x c l u s i v i t y , where the column elements of the v e c t o r s are d i s s i m i l a r , was a l s o found to be an e f f e c t i v e b a s i s of i d e n t i f y i n g c l u s t e r s . While the d e f i n i t i o n and i d e n t i f i c a t i o n of c l u s t e r s i n a t e x t data base was s a t i s f y i n g , no q u a n t i t a t i v e i n f o r m a t i o n about the closeness of elements w i t h i n the c l u s t e r or about the goodness of f i t could be obtained w i t h those d e f i n i t i o n s . The present paper, t h e r e f o r e , addresses these questions. D i s t r i b u t i o n of Uses The medical use data f i l e which i s being analyzed has been e x t r a c t e d from the mai Merck Index s i s t s of l i n e r e c o r d s f i r s t f i e l d i s a f o u r - d i g i t code number r e p r e s e n t i n g a unique chemical compound. The second f i e l d i s a l e t t e r c h a r a c t e r symb o l i z i n g a s p e c i f i c q u a l i f i e r a p p l i e d to the g i v e n medical use. Table I gives the 10 q u a l i f i e r s coded i n t h i s f a s h i o n . Note that the l e t t e r X denotes the set of uses without any q u a l i f i c a t i o n . The t h i r d f i e l d i s the name of the medical use expressed as an alphanumeric s t r i n g of c h a r a c t e r s . In f u r t h e r m a n i p u l a t i o n of the use data f i l e , the counting procedures and concepts of numerical l i n g u i s t i c s w i l l be used e x t e n s i v e l y . Although t h i s a p p l i c a t i o n of l i n g u i s t i c s to a data base not c o n t a i n i n g the u s u a l n a t u r a l language t e x t may seem odd, there are s e v e r a l advantages i n doing so. F i r s t , the methods already developed i n other contexts may be a p p l i e d without subs t a n t i a l reprogramming. Second, when the data base i s viewed l i n g u i s t i c a l l y , i t s s i m i l a r i t i e s or d i f f e r e n c e s w i t h other l i n g u i s t i c data can lead to hypotheses about how the data were generated. In other words, the r u l e s f o r naming medical uses w i l l be compared w i t h the naming r u l e s used i n o r d i n a r y language. T h i r d , there i s some i n d i c a t i o n that the problems of indexing and r e t r i e v i n g can be b e t t e r understood i n t h i s context i f the data base i s t r e a t e d as i f i t were normal t e x t used f o r b i b l i o graphic purposes; t h a t i s , e i t h e r as key words o r as a b s t r a c t s . To i n t r o d u c e important concepts, then, i t i s f i r s t necessary to d i s t i n g u i s h between a s p e c i f i c use name found i n the data base and a l l occurrences o f t h a t s p e c i f i c name. The s p e c i f i c name i t s e l f w i l l be c a l l e d a use type> w h i l e any s i n g l e occurrence of t h a t name w i l l be c a l l e d a use token. Thus, i f the use narcotic occurs 48 times i n the data base, then t h a t set could be d e s c r i b e d as 48 use tokens; o r , on the other hand, i t could be d e s c r i b e d as the use type narcotic having a frequency of 48 i n the data base. This usage i s very convenient f o r counting purposes. The data base contains 3433 chemical compounds. There are


4.

MARCUS E T AL.

Free-Text

Data

Bases

TABLE I

SPECIFIC QUALIFIERS GIVEN TO MEDICAL USES IN THE MERCK INDEX

X

N u l l Character

Ζ

A d d i t i o n a l Information

Η

Has been used

F

Formerly

A

A c t i v i t y , Properties

Ε

Experimental

I

Has been i n v e s t i g a t e d , has been s t u d i e d , i n v e s t i g a t i v e , has been t r i e d as

Ρ

Proposed as

R

Reportedly causes

S

O c c a s i o n a l l y a s , sometimes as


41

42

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

949 d i s t i n c t medical uses, or use types, mentioned i n the use f i l e . But s i n c e each compound may have more than one d i f f e r e n t use, the t o t a l number of use tokens i s expected t o be l a r g e r than the t o t a l number of compounds. In f a c t , there are 4848 use tokens i n the f i l e , which represents about 1.4 uses on the average f o r a t y p i c a l compound. Hence, each use type occurs on the average of about 5.1 times; that i s , there are over 5 use tokens for each t y p i c a l use type. S t a t i s t i c s l i k e those j u s t c i t e d — a v e r a g e number of uses per compound or average use tokens per t y p e — a p p l y t o the e n t i r e data base and g i v e some n o t i o n of gross p r o p e r t i e s . But more d e t a i l e d s t r u c t u r e s can be measured and w i l l g i v e a b e t t e r p i c ture of the naming r u l e s which l e d t o assignment of use names. In f a c t , more can be learned by p a r t i t i o n i n g the data s e t i n t o s m a l l e r subsets and c a l c u l a t i n g the p r o p e r t i e s of these d i v i s i o n s . To the extent t h a t thes whole f i l e , i t i s p o s s i b l a s i g n i f i c a n t a n a l y t i c a l o p e r a t i o n on the data. Use Combinations. Each compound i n the data base has one or more d i s t i n c t use names a s s o c i a t e d w i t h i t . For some compounds, the s e t of names may c o n t a i n two that are the same but whose ( l e t t e r ) q u a l i f i e r s are d i f f e r e n t . These d i f f e r e n t l y q u a l i f i e d use names w i l l be considered as d i f f e r e n t use names f o r the purpose of t h i s i n v e s t i g a t i o n . Thus, each compound can be assigned an i n t e g e r which denotes how many d i s t i n c t use tokens are a s s o c i a t e d w i t h i t . This i n t e g e r , which measures uses per compound, can a l s o be assigned t o each use token a s s o c i a t e d w i t h the given compound. In the data base, the number of uses per compound v a r i e s g e n e r a l l y up t o 6, but there are a s p e c i a l s e t of compounds which have 9 a s s o c i a t e d uses each. The f i r s t r e s t r u c t u r i n g of the data f i l e c o n s i s t e d , then, i n a s s i g n i n g a uses-per-compound i n t e g e r t o a l l use tokens and an assignment of a code number t o each use type. This l a t t e r assignment was made by l e t t i n g the most frequent use type have the lowest code number. The data f i l e was f u r t h e r reorganized by combining a l l use tokens (now represented by code numbers) for a given compound on the same l i n e i n ascending numerical order. Thus, the 4848-line f i l e was reduced t o a 3433-line f i l e . Only the use q u a l i f i e r data was removed. The r e s u l t i n g f i l e was then s o r t e d by uses per compound and s e c o n d a r i l y by the p a r t i c u l a r combination of uses. F i n a l l y , a counting program was used to e l i m i n a t e compound numbers and group together a l l s i m i l a r use combinations. This p a r t i t i o n i n g of the main f i l e i n t o subsets w i t h d i f f e r ent values of uses per compound (that i s , s i z e of the use combinat i o n s e t ) provided some s t a t i s t i c s on how m u l t i p l e uses are d i s t r i b u t e d . Table I I presents the summary data. I t shows how the number of compounds and use tokens are d i s t r i b u t e d over the uses/compound parameter. The column l a b e l e d "use types"


4.

MARCUS E T A L .

Free-Text

Data

43

Bases

TABLE I I

MULTIPLE USE TABLE SUMMARY DATA D i s t r i b u t i o n of Number of Compounds and Their Occurrence Over the Uses/Compound Parameter

Uses/ Compound

Number Of Compounds

Use Tokens (Occurrences)

Use Types (Descriptors)

D i s t i n c t Use Combinations

1

2447

2447

570

570

2

717

1434

414

391

3

187

561

231

133

4

60

240

140

46

5

5

25

21

5

6

4

24

20

4

9

13

117

10

2

3433

4848

TOTALS

1151


44

RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

gives the number of use types which appear i n each subset. Since the p a r t i t i o n i n g i s o v e r l a p p i n g i n use types, the numbers i n that column cannot be added t o give a t o t a l . The l a s t column i n d i cates how many d i s t i n c t l y d i f f e r e n t use combinations there are w i t h a given combination s i z e . For example, each compound w i t h 5 or 6 uses has a d i s t i n c t s e t of a s s o c i a t e d uses, w h i l e 9-use compounds have only 2 d i s t i n c t s e t s . Since use combinations are p a r t i t i o n e d i n a non-overlapping f a s h i o n , a sum of the column e n t r i e s gives 1151 as the t o t a l d i s t i n g u i s h a b l e number of use combinations. Thus, on the average, each use combination i s a s s o c i a t e d w i t h about 3 d i f f e r e n t compounds. In one v e r s i o n of the f i l e s o r t e d on use combination, a l l use s e t s beginning w i t h the same use are grouped together, r e g a r d l e s s of the s i z e of d i f f e r e n t combinations. This format permits comparison of use combinations c o n t a i n i n g the same p a i r s of uses. An e y e b a l l search, w i t t a b u l a t e the most frequen Table I I I i n c l u d e s a l l use p a i r s w i t h a frequency of occurrence g r e a t e r than 5. The p a i r names and frequency of occurrence are presented i n the t a b l e , as w e l l as the d i s t r i b u t i o n of these p a i r s over d i f f e r e n t s i z e combinations. For example, the use p a i r analgesia-sedative occurs three times as a p a i r but a l s o i n three t r i p l e combinations w i t h other uses: h y p n o t i c , n a r c o t i c , and s k e l e t a l muscle r e l a x a n t . The e n t r i e s i n Table I I I have been grouped by closeness o f r e l a t i o n . I n a taxonomic sense, the grouped p a r t s c o n s t i t u t e a c l u s t e r . The f i r s t e i g h t e n t r i e s represent an a n a l g e s i c - s e d a t i v e c l u s t e r , which i n c l u d e s the use names analgésie or sedative. The next four t a b l e e n t r i e s represent an a n t i s e p t i c - a s t r i n g e n t c l u s t e r . The cardiotonic-cardiac rate decrease p a i r i s i s o l a t e d and seems t o suggest synonymity. The next f i v e e n t r i e s are a c l u s t e r centered on the diuretic'-antihypertensive pair. Proceeding f u r t h e r down the t a b l e , one encounters a c l u s t e r based on a t r i p l e of uses: a d r e n o c o r t i c a l s t e r o i d , g l u c o c o r t i c o i d , a n t i - i n f l a m m a t o r y . The parasympathomimetic-miotic c l u s t e r occurs alone, f o l l o w e d by a c l u s t e r centered on sympathomimetic. Use-Oriented Data. To i n v e s t i g a t e r e l a t e d uses, the data base was r e s t r u c t u r e d t o p l a c e a l l uses a s s o c i a t e d w i t h a given compound together i n the same record (or e q u i v a l e n t l y on the same t e x t l i n e ) . But other i n f o r m a t i o n may be obtained by grouping together a l l tokens o f the same use type. I n a f i l e whose records c o n s i s t o f the compound code, the uses-per-compound parameter, the q u a l i f i e r code, and the use name, a simple alphanumeric s o r t on use name b r i n g s together a l l use tokens of the same type. The r e s u l t i n g f i l e can then be f u r t h e r s o r t e d on e i t h e r uses per compound or q u a l i f i e r code. From these sorted f i l e s , which s t i l l c o n t a i n compound codes, a summary t a b l e can be prepared by removing compound r e f e r e n c e s . I n such a t a b l e , each r e c o r d or t a b l e l i n e contains a count of a l l use tokens


4.

MARCUS E T AL.

Free-Text

Data

Bases

TABLE I I I

DOMINANT USE PAIRS

Uses/Compound

29 27 3 4 6 3 54 4 14 12 4 6

_3

4_

>5

12 5

1 1

14

5 4 1 3 3 7

10 1

P a i r Names

56 33 10 9

analgesic, a n t i p y r e t i c analgesic, narcotic analgesic, anesthetic a n a l g e s i c , antirheumatic

6 58 11

analgesic, sedative s e d a t i v e , hypnotic sedative, t r a n q u i l i z e r

26 13 6 8

antiseptic, antiseptic, antiseptic, astringent,

22

cardiotonic, cardiac rate decrease d i u r e t i c , smooth muscle relaxant d i u r e t i c , myocardial stimulant d i u r e t i c , antihypertensive diuretic, vasodilator a n t i h y p e r t e n s i v e , ganglion b l o c k i n g agent

1

1

12 1 1 2

1

22

15

7

22

14

7

21

7

14 7 12

14 10

Freq.

2

8

6

1

15

4

6

1

11

5

2

1

4 5 4

2 1 2

astringent disinfectant expectorant styptic

adrenocortical steroid, glucocorticoid adrenocortical steroid, anti-inflammatory

8

parasympathomimetic, miotic

6 6 6

sympathomimetic, v a s o c o n s t r i c t o r sympathomimetic, CNS stimulant sympathomimetic, decongestant


46

RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

(occurrences) of a given type (the frequency of t h a t type) together w i t h the use name. Separate columns then present counts of the numbers of use tokens appropriate t o each value of the secondary s o r t parameter. Table IV presents a sample of such a t a b u l a r p r e s e n t a t i o n f o r uses per compound. Note that the t a b l e has been sorted so that the most frequent uses occur f i r s t i n the t a b l e . The type of data f i l e d i s p l a y e d i n Table IV resembles a concordance i n the sense that i t shows occurrences of d i f f e r e n t words i n d i f f e r e n t subclasses or t e x t s . In most l i n g u i s t i c s t u d i e s , the columns represent d i f f e r e n t t e x t s , and the t a b l e e n t r i e s denote counts of word tokens i n those t e x t s . We w i l l use the d e s c r i p t i o n " t e x t " here to denote one of these columns even though no t e x t has been d e f i n e d ; r a t h e r , " t e x t " means a p a r t i t i o n i n g of the data. The e n t i r e t a b l e w i l l be c a l l e d a word d i s t r i b u t i o n t a b l e r i n g t o these p a r t i c u l a An i n s p e c t i o n of Table IV r e v e a l s that there appear t o be two d i s t i n c t i v e l y d i f f e r e n t c l a s s e s of use types as revealed by the d i s t r i b u t i o n of tokens w i t h the parameter uses/compound. A comparison of antimicrobial w i t h analgesic i s i n s t r u c t i v e . Antimicrobial has the l a r g e s t number of s i n g l e uses (201) i n the f i l e and has only 25 m u l t i p l e uses. I t a l s o describes no compounds w i t h more than three uses. This suggests e i t h e r t h a t compounds w i t h a n t i m i c r o b i a l p r o p e r t i e s are h i g h l y s p e c i f i c o r that the name i s a p p l i e d very s p e c i f i c a l l y . In c o n t r a s t , analgesic has only 55 s i n g l e uses, but 119 m u l t i p l e uses. I n f a c t , one compound w i t h a n a l g e s i c p r o p e r t i e s has s i x uses. T h i s suggests e i t h e r that compounds w i t h a n a l g e s i c p r o p e r t i e s a r e r a t h e r nons p e c i f i c and g e n e r a l i z e d i n t h e i r a c t i o n or t h a t the name analgesic r e f e r s t o a r a t h e r wide range of e f f e c t s . These examples appear t o suggest t h a t , by examining the d i s t r i b u t i o n of tokens w i t h the uses-per-compound parameter, the uses can be e i t h e r p a r t i t i o n e d i n t o s p e c i f i c and n o n s p e c i f i c c l a s s e s or described by a s p e c i f i c i t y parameter. Frequency D i s t r i b u t i o n . The word d i s t r i b u t i o n t a b l e prov i d e s not only the d i s t r i b u t i o n of word tokens among the v a r i o u s t e x t s but a l s o the t o t a l frequency of each word type. The t o t a l number of types or tokens having a given frequency can be d e t e r mined by simply counting types i n the d i s t r i b u t i o n t a b l e or summing the t o t a l number of tokens w i t h that frequency. The f u n c t i o n s which present the number of types or tokens of a given frequency are c a l l e d , r e s p e c t i v e l y , the type and token distribution functions. In general word s t a t i s t i c s , the type frequency d i s t r i b u t i o n i s a s t e e p l y decreasing f u n c t i o n of frequency; the s i n g l y o c c u r r i n g words c o n s t i t u t e between 25% and 40% o f the t o t a l number of types. A t frequencies greater than 30 o r 40 ( f o r a s m a l l corpus), the type d i s t r i b u t i o n f u n c t i o n i s zero a t many


4.

MARCUS E T AL.

Free-Text

Data

Bases

TABLE IV USE DISTRIBUTION FILE Frequency Sort f o r Uses Per Compound Uses/Compound 1 201 93 55 30 47 54 22 71 52 46 58 19 5 27 13 14 29 15 12 23 29 35 40 27 23 32 12 16 10 23 12 5 22 14 14 21 20 23 2

2

3

24 55 82 72 36 23 59 10 10 20 10 32 33 32 22 29 16 20 29 8 11 8 1 5 10 1 24 7 13 9 12 3 7 14 7 5 3 1 1

1 24 21 15 22 7 4 1 3 7 2 17 12 3 12 4 5 8 6 1 4 1 4 5 2 1 6 6 2 6 16 1 2 4

4

6

1 2 3 9 2 2

1

1 1

2 1

13

1

2

10

1 1

2

13

6 4

2

3 1 1 1

2

1

13

3 1 3 1 9 2 1 7

2

7

2 1

2

Use Name

9

2

1 14

Freq.

5

226 175 174 120 117

82 79 73 70 69 66 62 56 51 50 48 48 46 45 44 41 39 39 38 38 38 35 35 31 31 30 30 27 26 24 24 24

antimicrobial antiseptic analgesic sedative diuretic

antineoplastic anesthetic tranquilizer antihistaminic astringent antipyretic an t ihyρ ertens i v e expectorant sympathomimetic anthelmintic carminative narcotic antimalarial antitussive antituberculous x-ray c o n t r a s t medium cathartic CNS s t i m u l a n t anesthetic l o c a l cardiotonic vasodilator counterirritant laxative adrenocortical steroid smooth muscle r e l a x a n t amebicide ganglion b l o c k i n g agent parasympathomimet i c v a s o d i l a t o r coronary antifungal estrogenic myocardial s t i m u l a n t

American Chemical Society Library 1155 16th St. N. W. In Retrieval of Medicinal Chemical Washington, D. C. Information; 20096 Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

48

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

frequencies and has v a l u e s between 1 and 5. At l a r g e frequencies the f u n c t i o n i s zero except a t a very few f r e q u e n c i e s . The nature of token frequency d i s t r i b u t i o n f u n c t i o n s i s somewhat d i f f e r e n t , even though f o r l a r g e frequencies they are a l s o nonzero a t the same s e l e c t e d f r e q u e n c i e s . For purposes of a n a l y s i s and modeling, i t i s b e t t e r t o work w i t h cumulative d i s t r i b u t i o n s . Cumulative d i s t r i b u t i o n s a r e formed by summing the frequency f u n c t i o n s f o r a l l f r e q u e n c i e s l e s s than a c e r t a i n value and d i v i d i n g by the t o t a l sum. Cumulative d i s t r i b u t i o n s , when p l o t t e d as a f u n c t i o n of frequency, are S-shaped (ogive) curves, ranging from 0 a t zero frequency t o 1 a t l a r g e frequency. Numerical l i n g u i s t s have made many attempts t o d e s c r i b e or model the word frequency d i s t r i b u t i o n . One of the models which has e x t e n s i v e use i s the log-normal d i s t r i b u t i o n . This model can be d e f i n e d i n terms of the cumulative normal p r o b a b i l i t y d i s t r i b u t i o n . The log-norma lative distribution i the r e s u l t i n g curve conforms t o a normal d i s t r i b u t i o n . The l o g normal d i s t r i b u t i o n d e s c r i b e s many l o n g - t a i l e d d i s t r i b u t i o n s i n l i n g u i s t i c s c i e n c e ; i n p a r t i c u l a r , the d i s t r i b u t i o n s of word l e n g t h and sentence l e n g t h ( 3 ) . Log-normal d i s t r i b u t i o n s are a l s o found i n econometrics and other s o c i a l phenomena ( 4 ) . What gives the log-normal d i s t r i b u t i o n p a r t i c u l a r i n t e r e s t i n word frequency analyses i s t h a t i f the type frequency d i s t r i b u t i o n i s log-normal, then the token frequency d i s t r i b u t i o n i s a l s o log-normal. In p r a c t i c a l terms, t h e type and token cumulative frequency d i s t r i b u t i o n s may be t e s t e d f o r l o g - n o r m a l i t y by p l o t t i n g these f u n c t i o n s on normal p r o b a b i l i t y paper w i t h the l o g a r i t h m of f r e quency as the a b s c i s s a . When t h i s t e s t was a p p l i e d t o the medical use type and token frequency d i s t r i b u t i o n s , the log-normal model was found t o d e s c r i b e both d i s t r i b u t i o n s very w e l l over two orders of magnitude. F i g u r e 1 shows t h i s r e l a t i o n s h i p f o r the use type d i s t r i b u t i o n and a l s o f o r the c l u s t e r type d i s t r i b u t i o n . The c l u s t e r types i n v o l v e a d i f f e r e n t counting than use types, so t h a t two c l u s t e r s a r e the same i f a l l t h e i r components are i d e n t i c a l . The t o t a l number of c l u s t e r types i s simply the sum of the l a s t column i n Table I I . P l o t t e d f o r comparison on F i g u r e 1 are two cumulative d i s t r i b u t i o n s i n which the frequency i s d i s t r i b u t e d normally. Note that the normal curves cannot s a t i s f a c t o r i l y f i t the long t a i l s of the log-normal d i s t r i b u t i o n s . While the normal curves have been chosen somewhat a r b i t r a r i l y t o f i t the observed d i s t r i b u t i o n s a t frequencies of 1 and 22, there i s no c o n s i s t e n t way i n which t o f o r c e a l o g a r i t h m i c curve t o f i t a l i n e a r one. P l o t t e d i n F i g u r e 2 a r e the cumulative type frequency d i s t r i b u t i o n s f o r two subsets of the data base. The s i n g l e use curve corresponds t o a l l compounds w i t h a s i n g l e use. The X curve corresponds t o a l l uses w i t h the X ( n u l l ) q u a l i f i e r ; i . e . , a l l uses which are u n q u a l i f i e d . These subsets a r e a l s o seen t o


MARCUS ET AL.

Free-Text

Data

Bases

Figure 1. Log-normality of type distribution. Points are plotted on a log frequency scale; curves represent a normal, rather than log-nor mal, distribution in frequency. Cluster types, O ; use types, ·. Normal curve vs. frequency fit to data: clusters, ; uses, . 10

Log φ Frequency

Figure 2. Log-normality of data base subsets. Points are plotted on a log frequency scale; curve represents a normal, rather than log-nor mal, distribution in frequency. Single uses, ·; X uses (no qualifier), O . Normal curve vs. fre quency fit to single-use data. l0


50

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

be log-normally d i s t r i b u t e d . The f a c t that the use frequency d i s t r i b u t i o n conforms so c l o s e l y to the log-normal was unexpected, inasmuch as the s t a t i s t i c a l " r u l e s " which operate i n n a t u r a l language t e x t would not n e c e s s a r i l y be assumed to operate f o r a corpus of terms derived from a h i g h l y t e c h n i c a l area of s p e c i a l t y such as pharmacology. However, the f i n d i n g suggests that the Merck Index medical use data base does not d i f f e r s u b s t a n t i a l l y i n i t s l i n g u i s t i c prope r t i e s from n a t u r a l language data bases. This r e s u l t i s c o n s i s tent w i t h s t u d i e s of d i s e a s e i d e n t i f i c a t i o n , which a l s o i n v o l v e s a p r o f e s s i o n a l code of terms (3). S p e c i f i c i t y . Several attempts were made t o develop a nonparametric s t a t i s t i c a l d e s c r i p t i o n of the s p e c i f i c i t y q u e s t i o n , based on techniques from numerical l i n g u i s t i c s . I t was hoped that s e t s of very s p e c i f i n a t u r a l l y emerge i n th simple parametric s t a t i s t i c d i d f a r b e t t e r than nonparametric ones: the mean uses per compound f o r each use. The mean uses per compound f o r the e n t i r e data base (excluding 13 compounds w i t h 9 uses, which represent a s p e c i a l case) i s c a l c u l a t e d to be 1.739. The d i f f e r e n c e between the p o p u l a t i o n mean and the mean f o r any use can be t e s t e d f o r s i g n i f i c a n c e by comparison to the standard d e v i a t i o n . For a l a r g e use frequency, N, the mean uses per compound, û", i s d i s t r i b u t e d normally w i t h a s t a n dard d e v i a t i o n given by σ = (.899/N)

15

F i g u r e 3 shows the d i s t r i b u t i o n of values of û f o r f r e quencies of 6 or g r e a t e r . The curves r e p r e s e n t i n g ±2o and ±3σ l i m i t s about the p o p u l a t i o n mean are a l s o shown. From t h i s f i g u r e , 25 uses were found f o r which û d i f f e r e d by 3o from the p o p u l a t i o n mean. Eight uses have s i g n i f i c a n t l y low û: seventeen uses have s i g n i f i c a n t l y h i g h û. L i s t s of these uses w i t h values of û and standard d e v i a t i o n s are given i n Table V. The parameter û was a l s o c a l c u l a t e d f o r the set of compounds corresponding to each use q u a l i f i e r . The s e t s described by q u a l i f i e r s X and Ζ both showed û v a l u e s not s i g n i f i c a n t l y d i f f e r e n t from the t o t a l p o p u l a t i o n . The H and F sets both showed higher than average û. Except f o r the R q u a l i f i e r , the remaining sets a l l had û l e s s than average. Table VI shows these r e s u l t s w i t h combined s e t s to increase s i g n i f i c a n c e . S i g n i f i c a n c e i s denoted by c a l c u l a t e d σ d e v i a t i o n s from the p o p u l a t i o n mean. In t h i s t a b l e , X and Ζ are average, Η and F combined have s i g n i f i c a n t l y h i g h e r uses/compound, and the remaining combination has s i g n i f i c a n t l y lower uses/compound. Thus X and Ζ compounds seem t y p i c a l of the average com pound i n the data base, w h i l e a compound w i t h an Η or F q u a l i f i e r has s i g n i f i c a n t l y more uses than average. This r e s u l t suggests


Figure 3. Distribution of mean uses per compound. Outliers are identified in Table V. Poputotion mean, ; 2 σ deviation, ; 3 σ deviation, .


52

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

TABLE V

USES HAVING SIGNIFICANT SPECIFICITY PARAMETER

Mean Uses Per Compound

Use Name

D e v i a t i o n From P o p u l a t i o n Mean (In Q ' S )

Low Mean Uses/Compound antimicrobial antineoplastic antihistaminic antituberculous x-ray c o n t r a s t medium anesthetic l o c a l estrogenic anemia i r o n d e f i c i e n c y

1.11 1.14 1.200 1.227 1.024 1.143 1.042 1.000

4.8 3.6 4.8 3.7 3.6 3.1

High Mean Uses/Compound diuretic antipyretic expectorant vasodilator counterirritant smooth muscle r e l a x a n t myocardial s t i m u l a n t antirheumatic glucocorticoid mydriatic CNS depressant decongestant diaphoretic carbonic anhydrase inhibitor dermatides dermatoses sudorific

2.026 2.321 2.375 2.211 2.400 2.806 3.083 2.591 2.500 2.750 2.700 2.700 2.700

3.3 4.5 5.0 3.1 4.1 6.3 6.9 4.2 3.3 3.7 3.2 3.2 3.2

2.778 2.857 3.167 3.167

3.3 3.1 3.7 3.7


4.

MARCUS E T A L .

Free-Text

Data

Bases

53

e i t h e r t h a t compounds w i t h many uses tend t o have more " o l d " uses or t h a t " o l d " use names tended t o be l e s s s p e c i f i c . The remain ing q u a l i f i e r group has s i g n i f i c a n t l y fewer uses than average, which suggests e i t h e r that compounds w i t h "new" medical uses tend to be l a b e l e d w i t h only one use or that "new" d e s c r i p t o r s are more s p e c i f i c . Discussion We have d e s c r i b e d i n the preceeding s e c t i o n four d i f f e r e n t cuts through the hyperspace of medical uses and t h e i r frequency i n the Merck Index data base. By " c u t " we r e f e r by analogy t o a p r o j e c t i o n of the hyperspace onto one or more r e c o g n i z a b l e orthogonal C a r t e s i a n axes. I n t h i s s e c t i o n we w i l l d i s c u s s what we have learned from these four c u t s . We w i l l a l s o approach, but c e r t a i n l y not exhaust e f f e c t space) w i t h th ture of compounds having these uses (the s t r u c t u r e space). Use Frequencies. One cut through the data base i s t o p l o t the frequency d i s t r i b u t i o n of the medical uses and v a r i o u s major subsets t h e r e o f . We show i n F i g u r e s 1 and 2 t h a t the cumulative d i s t r i b u t i o n of both use types ( d e s c r i p t o r s ) and use tokens (occurrances) i s log-normal. Two major subsets of the data base a l s o have a log-normal d i s t r i b u t i o n . This type of d i s t r i b u t i o n shows that the vocabulary of t h i s unique t e x t i s , c o n t r a r y t o what might have been expected, no d i f f e r e n t from any other t e x t which has been considered by numerical l i n g u i s t s . The log-normal d i s t r i b u t i o n found by numerical l i n g u i s t s a p p l i e s not o n l y t o v o c a b u l a r i e s i n v o l v i n g disease i d e n t i f i c a t i o n , but a p p l i e s t o a number of d i f f e r e n t n a t u r a l language data bases. Mean Uses/Compound. Another c u t through the data base has been made i n terms of mean uses per compound. This parameter has been p l o t t e d a g a i n s t frequency i n F i g u r e 3. I t i s t h i s f i g u r e which permits a f i r s t s t a t i s t i c a l e v a l u a t i o n of the s i g n i f i c a n c e of c l u s t e r s o c c u r r i n g i n t e x t data bases. L i n e s corresponding to two and three standard d e v i a t i o n s (σ) from the mean are drawn i n t o F i g u r e 3, and the v a r i o u s o u t l i e r s on e i t h e r s i d e are l i s t e d i n Table V. The reasons why p a r t i c u l a r uses have low or high mean uses/compound are perhaps best l e f t t o pharmacologists. I t i s r e a d i l y apparent, however, even t o nonpharmacologists, that the uses l i s t e d under low mean uses/compound are h i g h l y s p e c i f i c i n a l i n g u i s t i c , as w e l l as s t a t i s t i c a l , sense. Four of them a r e " a n t i - , " and the other stem of these f o u r words i s a h i g h l y spe c i f i c concept: - m i c r o b i a l , - n e o p l a s t i c , - h i s t a m i n i c , and -tuberculous. Another three uses under low mean uses/compound depend on d i s t i n c t l y chemical p r o p e r t i e s ; these are x-ray con t r a s t medium, anemia i r o n d e f i c i e n c y , and e s t r o g e n i c . Uses l i s t e d under h i g h mean uses/compound seem t o be of two k i n d s . One of


54

RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

these i s the k i n d of use such as d i u r e t i c , v a s o d i l a t o r , smooth muscle r e l a x a n t , myocardial s t i m u l a n t , a n t i p y r e t i c , and expec t o r a n t , which are present i n the dominant use p a i r s , o r c l u s t e r s , l i s t e d i n Table I I I and which w i l l be discussed l a t e r on i n t h i s s e c t i o n . O b v i o u s l y , such compounds would, by t h e i r presence i n these c l u s t e r s , be expected to have more uses per compound than the average. Other uses i n t h i s h i g h mean uses/compound r u b r i c are those use d e s i g n a t i o n s which are o l d e r and perhaps somewhat i m p r e c i s e by more modern d e f i n i t i o n s . This p o i n t w i l l become apparent i n the next paragraph when we d i s c u s s q u a l i f i e r s . Q u a l i f i e r s . A t h i r d cut through the hyperspace was taken on the b a s i s of the q u a l i f i e r s which are defined i n Table I . I t was shown i n Figure 2 that a major subset defined by the use q u a l i f i e r X (a n u l l c h a r a c t e r ) e s s e n t i a l l y followed the same l o g normal d i s t r i b u t i o n a Table V I shows that bot s e t s do not have a s t a t i s t i c a l l y s i g n i f i c a n t d e v i a t i o n from the p o p u l a t i o n mean. However, uses having q u a l i f i e r s Η or P, which represent former uses, have h i g h l y s i g n i f i c a n t (5.3 o) greater mean uses/compound, a f a c t which was a l l u d e d t o i n a d i f f e r e n t context i n the previous paragraph. F o r t u n a t e l y f o r t h i s s t a t i s t i c a l view of the Merck Index data base, the use q u a l i f i e r s A, Ρ, T, P_, P, and S, which represent experimental compounds not yet i n g e n e r a l use, together have fewer mean uses/compound w i t h s t a t i s t i c a l l y h i g h s i g n i f i c a n c e (4.5 σ ) . The d i s c u s s i o n i n t h i s s e c t i o n i n d i c a t e s that not a l l subsets of the data base have the same d i s t r i b u t i o n and that use q u a l i f i e r s , such as those l i s t e d i n Table I , which were f r e q u e n t l y used i n the e i g h t h e d i t i o n of the Merck IndeXj can d i s t i n g u i s h between d i f f e r e n t kinds of med i c a l uses i n a s t a t i s t i c a l l y s i g n i f i c a n t manner. Use C o r r e l a t i o n s . A f o u r t h cut through the medical usefrequency hyperspace i s the one which i n v o l v e s the d e r i v a t i o n of dominant use p a i r s . Table I I I l i s t s these use p a i r s as a f u n c t i o n o f uses/compound. Table I I I a l s o l i s t s the frequency of i n d i v i d u a l use p a i r s . I t was noted e a r l i e r that the use p a i r s i n Table I I I have been grouped by closeness of r e l a t i o n and that i n a taxonomic sense, the grouped p a r t s c o n s t i t u t e a c l u s t e r . The seven c l u s t e r s i d e n t i f i e d are shown i n Table I I I as sepa r a t e d groups of use p a i r s . I t i s t h i s p a r t i c u l a r cut through the data base which suggests the p o s s i b i l i t y of e x t r a p o l a t i o n . Many compounds which have one use contained i n a c l u s t e r might w e l l be examined p h a r m a c o l o g i c a l l y or c l i n i c a l l y t o see whether that compound a l s o has one o r more of the other uses i n that cluster. The same reasoning which suggests t h i s p o s s i b i l i t y of e x t r a p o l a t i o n a l s o leads t o the c o n c l u s i o n that i t i s t h i s p a r t i c u l a r cut through the medical use hyperspace which leads i n t o a s s o c i a t i o n s w i t h the chemical s t r u c t u r e hyperspace.



1106 255

H + F

+ E + I + P + R+

D e f i n i t i o n s i n Table 1. No 9-use compounds i n c l u d e d .

4731

1062

Ζ

T o t a l Uses**

2308

X

S

No. of Uses

Use Q u a l i f i e r *

1.739

1.490

1.881

1.711

1.711

Mean Uses/Compound

QUALIFIER SUBSETS OF USE TOKENS

TABLE VI

4.5

5.3

1.0

1.5

T

(In Q s )

D e v i a t i o n From P o p u l a t i o n Mean

56


Ideally, each of the uses i n Table I I I can be associated with a particular chemical structure. In previous work (1) we have shown that such associations can be made not only by pharmacolog i s t s , working from laboratory data and current theories, but also by text manipulation of the kind which we have described here. Thus for example, the effects of compounds i n the sympathomimetic cluster have been ascribed to the presence of the ethylamine grouping, while the effect of compounds i n the parasympathomimetic cluster have been ascribed to compounds containing the ammonium ion structure or its phosphorus analog. The compounds in the adrenocortical steroid cluster have similar chemical structures by d e f i n i t i o n . The cardiotonic cluster contains natural products related to those found i n digitalis and containing similar side chain structures. The antiseptic-astringent cluster contains recognizably caustic compounds. It i s hoped that similar association b mad i th d gesic clusters. The l a t t e r of uses and perhaps those with the greatest variety of d e f i n i tions. It would be interesting to be able to resolve these definitions on the basis of chemical structure as well as numerical l i n g u i s t i c s . Abstract Heuristic clustering methods for text data have been applied to a data base describing chemical compounds with medical uses. Clusters are sets of chemical compounds related by similarity of both chemical structure and a c t i v i t y . Previously discovered clusters in a free-text data base taken from the Merck Index could be described by a 2 x 2 activity-structure matrix. A s t a t i s t i c a l examination of the activity descriptors (medical uses) reveals them to have a log-normal distribution over two frequency decades. Log-normal distributions have been found by workers in other disciplines to be characteristic of random selection from a set of items with fixed limits. An investigation of compounds with more than one medical use revealed that the dominant clusters are labeled with the descriptors analgesic/ sedative, antiseptic/astringent, diuretic/antihypertensive, adrenocortical steroid, parasympathomimetic, and sympathomimetic. Chemical similarities are involved in those clusters based on multiple medical uses as well as in the case of clusters based on the 2 x 2 activity-structure matrix, which involved computer searches on single activity descriptors only. Consideration of multiple medical use clustering assures comprehensiveness and supplements any weakness of purely heuristic searching. The existence of these chemical similarities permits the exploration of new or previously unreported uses of chemical compounds by computer manipulation of text data.


4. MARCUS ET AL.

Free-Text Data Bases

57

Literature Cited 1. Marcus, R. J. and Gloye, Ε. Ε., "Real-Time I n t e r r o g a tion of Chemical D a t a , " J. Chemical Documentation 11, 163-7 (1971). 2. Marcus, R. J., Gloye, Ε. E., and F l o r a n c e , E . T., "Computer Search of a Free-Text Data Base as a Tool f o r Investi gating S t r u c t u r e - E f f e c t R e l a t i o n s h i p s , " Computers and Chemistry 1, 235-241 (1977). 3. Herdan, G., "The Advanced Theory of Language as Choice and Chance," S p r i n g e r - V e r l a g , New York, N.Y., 1966. 4. M o n t r o l l , E . W. and Badger, L . W . , " I n t r o d u c t i o n to Q u a n t i t a t i v e Aspects of S o c i a l Phenomena," pp. 110-120, Gordon and Breach, New Y o r k , N.Y., 1975. RECEIVED August 29, 1978


5 CHEMLINE: a Chemical Structure Search Key to Biological Information MELVIN L. SPANN, DONALD J. HUMMEL, ROBERT J. SCHULTHEISZ, SHARON L. VALLEY, and DONALD F. WALKER, JR. Toxicology Information Program, National Library of Medicine, Bethesda, M D 20014 This paper demonstrates the ability of an o n - l i n e chemical d i c t i o n a r y file in a c c e s s i n g computerized files c o n t a i n i n g biological As i n d i c a t e d in Figure 1, CHEMLINE, the N a t i o n a l L i b r a r y o f Medicine's (NLM) interactive, o n - l i n e d i c t i o n a r y file, can be viewed as the focal p o i n t f o r the chemical searching of the Library's on-line literature retrieval s e r v i c e s (1). In a d d i t i o n , CHEMLINE provides a l i n k a g e through a "Locator" d e s i g n a t i o n t o other files c o n t a i n i n g i n f o r m a t i o n r e l e v a n t t o h e a l t h and environmental concerns. An example of the latter is the Environmental P r o t e c t i o n Agency's Toxic Substances C o n t r o l Act (TSCA) Inventory Candidate List. Substances appearing on the Candidate List are referenced w i t h the EPATSCALIST Locator code. This d i s c u s s i o n will be limited t o CHEMLINE's use in connection w i t h the TOXLINE and RTECS files, which are available t o over 800 institutions that subscribe t o NLM's o n - l i n e s e r v i c e s . CHEMLINE CHEMLINE (CHEMical d i c t i o n a r y on-LINE) i s a f i l e of chemical d e s c r i p t o r s created by NLM's Toxicology Information Program i n c o l l a b o r a t i o n w i t h Chemical A b s t r a c t s Service (CAS). This f i l e contains n e a r l y 500,000 chemical substance names r e p r e s e n t i n g over 246,000 unique substances. Because of CHEMLINE's unique f i l e d e s i g n , i t has c a p a b i l i t i e s which support both f u l l s t r u c t u r e and substructure searching. F i g u r e 2 shows a t y p i c a l CHEMLINE u n i t r e c o r d . The f o l l o w i n g s e c t i o n h i g h l i g h t s the c h a r a c t e r i s t i c s of the f i e l d s w i t h i n t h i s r e c o r d , a l l of which are d i r e c t l y searchable (up t o 39 characters) and p r i n t a b l e . RN i s the Chemical A b s t r a c t s Service (CAS) R e g i s t r y Number. This u n i q u e l y assigned number of up t o nine d i g i t s appears i n a hyphenated format without l e a d i n g zeros.

This chapter not subject to U.S. copyright. Published 1978 American Chemical Society In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

5.

SPANN E T A L .

Figure 1.

RN MF NI NI SY SY SY SY SY SY SY WL NR RS RE CL MH fiH MΚ LO LO LO RC

59

CHEMLINE

NLM's CHEMLINE

file

- 553-24-2 - C15-H16-M.CL-H - C I . BASIC RED 5, MONOHYDROCHLORIDE ( 8 C I ) - 2,8-PHENAZINEDIAMINE, N(8),N(8),3-TRIMETHYL-.M0N0HYnR0CHL0RIDE (9CI) - C I . 50040 - AM I NOD IMETHYLAMI N0T0LUAÎ ι1NOZI NE HYDROCHLORIDE - NEUTRAL RED - TOLUYLENE RED - NEUTRAL RED W - C I . BASIC RED 5 - NUCLEAR FAST RED (BASIC DYE) - Τ C6C6 BN I N J E FZ L N l S l &6H - 3 - 6,6,6 - C4N2-C6-C6 - NC2NC2 - PHENAZINES / ( 6 9 - 7 " ) - DIMETHYLAMIi-iES (73-74) - NEUTRAL RED / (MINOR IM 7 5 ) - TOXLINE - MEDLARS - EPATSCALIST - R089-4926

HCL Figure 2.

CHEMLINE

record


60

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

MF i s the molecular formula of the compound expressed i n the H i l l convention. T h i s means t h a t , f o r organic compounds, the number of carbon atoms i s c i t e d f i r s t ; t h i s i s f o l l o w e d by the number of hydrogen atoms; and then a l l other elements are c i t e d i n a l p h a b e t i c a l order. For i n o r g a n i c compounds, a l l elements occur i n a l p h a b e t i c a l order. Nl i s the Chemical A b s t r a c t s (CA) p r e f e r r e d index name, i . e . , the s y s t e m a t i c names used i n the Chemical Substance and Formula Indexes of CA. The 8CI and 9CI d e s i g n a t i o n s f o l l o w i n g the two N l names f o r t h i s record i n d i c a t e that the name comes from the E i g h t h C o l l e c t i v e Index P e r i o d (1967-71) and the N i n t h C o l l e c t i v e Index P e r i o d (1972-76) of CA. One can see from t h i s example that the 9CI nomenclature i s c h e m i c a l l y more standardized than the 8CI nomenclature. SY i n d i c a t e s the synonymous names that CAS has on r e c o r d f o r a chemical substance. W i t h i uninverted chemical names mental names, as w e l l as company code numbers f o r chemicals. WL i s the Wiswesser L i n e N o t a t i o n (WLN) f i e l d . The WLN i s a unique and unambiguous r e p r e s e n t a t i o n of a chemical s t r u c t u r e diagram using a l i n e a r arrangement of 36 alphanumeric c h a r a c t e r s , 3 s p e c i a l characters and the blank. There are approximately 10,000 CHEMLINE records that c o n t a i n WLNs. The next s e t of f i e l d s (NR through CL) i n t h i s r e c o r d cont a i n s r i n g i n f o r m a t i o n . Since about 80% of the substances w i t h i n CHEMLINE c o n t a i n r i n g systems, one can see the importance of i n c o r p o r a t i n g t h i s i n f o r m a t i o n i n t o a chemical s t r u c t u r e search system, e s p e c i a l l y f o r substructure searching. NR p r o v i d e s the number of component r i n g s w i t h i n each unique r i n g system i n a chemical substance. As seen i n the diagram, there are 3 r i n g s w i t h i n the phenazine r i n g system. RS g i v e s the s i z e of the component r i n g s w i t h i n each unique r i n g system. For m u l t i p l e r i n g systems, r i n g s i z e s are c i t e d from s m a l l e s t t o l a r g e s t . I n the phenazine r i n g , the r i n g s i z e s are 6,6,6. RE, the Ring Elemental a n a l y s i s f i e l d , contains the molecular formula f o r each component r i n g w i t h i n a unique r i n g system. I n t h i s f i e l d , the element count begins w i t h carbon and a l l other elements (excluding hydrogen) f o l l o w i n a l p h a b e t i c a l order. For a m u l t i p l e - r i n g system, the r i n g elemental a n a l y s i s i s given f i r s t i n order of r i n g s i z e and secondly i n ascending order of the carbon atom count f o r i n d i v i d u a l r i n g s when r i n g s i z e s are equal. Therefore, the RE f i e l d f o r B a s i c Red 5 i s C4N2-C6-C6. Note that t h i s f i e l d i s a l p h a b e t i c a l and does not n e c e s s a r i l y provide the order of occurrence of the component r i n g s w i t h i n a r i n g system.


5.

SPANN

E T AL.

61

CHEMLINE

CL i s the Component Line formula f i e l d which provides a t o p o l o g i c a l d e s c r i p t i o n of those r i n g s c o n t a i n i n g between 4 and 8 atoms t h a t a l s o c o n t a i n two or more non-carbon atoms. The r u l e s f o r d e r i v i n g a Component Line formula are as f o l l o w s : 1. 2.

3.

S t a r t a t the e a r l i e s t a l p h a b e t i c non-carbon atom; proceed around the r i n g i n the d i r e c t i o n t h a t provides the s h o r t e s t path t o the next non-carbon atom; and where contiguous atoms r e p e a t , g i v e only the atom and i t s number of occurrences.

Thus, the example shown has the Component Line formula NC2NC2. MH o r the MeSH heading f i e l d contains terms from NLM's M e d i c a l Subject Headings (MeSH) c o n t r o l l e d vocabulary. There are approximately 4,00 LO contains the LOCATO of i n f o r m a t i o n or c i t a t i o n s r e l e v a n t t o the s p e c i f i c chemical substance r e t r i e v e d . CHEMLINE p r e s e n t l y contains l o c a t o r s t o MEDLARS, TOXLINE and EPATSCALIST. A l o c a t o r t o RTECS w i l l soon be added. RC contains the alphanumeric code assigned t o the approx i m a t e l y 33,000 compounds appearing on the EPA TSCA Inventory Candidate L i s t of Chemical Substances. This code should be reported t o EPA, along w i t h the CAS R e g i s t r y Number, t o s a t i s f y TSCA r e p o r t i n g requirements. In a d d i t i o n t o the i n f o r m a t i o n f i e l d s l i s t e d i n F i g u r e 2, there are s e v e r a l searchable but n o n - p r i n t a b l e f i e l d s w i t h i n CHEMLINE records that do not appear i n t h i s f i g u r e . Only two of these f i e l d s which are p a r t i c u l a r l y u s e f u l i n performing sub s t r u c t u r e searches w i l l be d i s c u s s e d . The f i r s t i s the NF or Name Fragment f i e l d . Name fragments are generated by computer programs from the NI (Type 1 name) and SY (Synonym) f i e l d s by breaking a name on hyphens, c o l o n s , enclosures and b l a n k s . Each uniquely o c c u r r i n g character s t r i n g becomes a searchable e n t i t y . Thus, the name fragments f o r the second N l name shown i n Figure 2 are: 2,8 phenazinediamine Ν 8

3 Trimethyl Monohydrochloride

The contents of the second fragment f i e l d , FF, are the molecular formula fragments which are d e r i v e d by breaking the molecular formula on hyphens and p e r i o d s . Excluding hydrogen atoms, each elemental symbol and i t s count are s t o r e d as search able e n t i t i e s . I n a d d i t i o n , the elemental symbol without a count i s saved f o r the heteroatoms ( n i t r o g e n , oxygen, phosphorus and s u l f u r ) and the halogens ( f l u o r i n e , c h l o r i n e , bromine and i o d i n e ) .


62

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

This f e a t u r e permits one to search f o r any ( u n s p e c i f i e d ) number of hetero or halogen atoms i n a chemical compound, or f o r a s p e c i f i c number of these atoms. As an example, searching w i t h the formula fragment S would r e t r i e v e a l l substances having a t l e a s t one s u l f u r atom i n i t s s t r u c t u r e ; whereas formula fragment SI would r e t r i e v e records f o r substances having o n l y one s u l f u r atom. The molecular formula fragments r e t a i n e d f o r B a s i c Red 5 a r e : C15 N4 CLl

Ν CL

TOXLINE TOXLINE (TOXicolog c o l l e c t i o n of computerize t o x i c o l o g y 425,000 r e f e r e n c e s (as of 1978) t o p u b l i s h e d human and animal t o x i c i t y s t u d i e s , e f f e c t s of environmental chemicals and p o l l u t a n t s , and adverse drug r e a c t i o n s . TOXLINE covers the p u b l i s h e d l i t e r a t u r e from 1974 forward. Older i n f o r m a t i o n (380,000 r e f e r ences) can be found i n TOXBACK, which i s searchable i n an o f f l i n e mode. TOXLINE/TOXBACK are made up of component f i l e s that come from the major secondary l i t e r a t u r e sources, as w e l l as s p e c i a l i z e d sources. The c o l l e c t i o n of b i b l i o g r a p h i c c i t a t i o n s a r e obtained e i t h e r through p r o f i l i n g c e r t a i n f i l e s or s e l e c t i n g com p l e t e s p e c i a l t y f i l e s . The sources and extent of coverage of component f i l e s a r e : C h e m i c a l - B i o l o g i c a l A c t i v i t i e s (CBAC) Sections 1-5 from 1965, S e c t i o n s 62-64 from 1975 and Sections 8, 59 and 60 from 1975; T o x i c i t y B i b l i o g r a p h y (TOXBIB) from 1968; A b s t r a c t s on H e a l t h E f f e c t s of Environmental P o l l u t a n t s (HEEP) from 1972; I n t e r n a t i o n a l Pharmaceutical A b s t r a c t s (IPA) from 1970; P e s t i c i d e s A b s t r a c t s (PESTAB, f o r m e r l y HAPAB) from 1966; Environmental Mutagen Information Center (EMIC) f i l e from 1960; Environmental Teratology Information Center (ETIC) from 1950; and the T o x i c M a t e r i a l s Information Center (TMIC) from 1971-1975. In a d d i t i o n , TOXBACK contains a t e r a t o l o g y f i l e (TERA) covering the years 1960-1974 and a s p e c i a l c o l l e c t i o n of l i t e r a t u r e con cerning the h e a l t h e f f e c t s of p e s t i c i d e s , the Hayes F i l e , which covers the p e r i o d from 1940-1966. As i n d i c a t e d i n F i g u r e 3, a l l of these f i l e s are arranged so that they can be searched simultaneously i n response t o a search query; they can a l s o be 'searched independently through use of the f i l e acronym. Each r e c o r d i n TOXLINE contains a f u l l b i b l i o g r a p h i c c i t a t i o n , most have a b s t r a c t s and/or i n d e x i n g terms and CAS R e g i s t r y Numbers. On-line r e t r i e v a l of TOXLINE records i s u s u a l l y accomplished by f r e e - t e x t searching of TEXTWORDS (TW) or


5.

SPANN E T A L .

63

CHEMLINE

keywords found i n t i t l e s , index f i e l d s or a b s t r a c t s of c i t a t i o n s . Other searchable elements i n c l u d e author name, CAS R e g i s t r y Number, secondary source i d e n t i f i c a t i o n , language, year of p u b l i c a t i o n , and j o u r n a l coden. RTECS While TOXLINE/TOXBACK are b i b l i o g r a p h i c r e t r i e v a l s e r v i c e s , the t h i r d f i l e t o be b r i e f l y d e s c r i b e d i n t h i s paper can be considered an o n - l i n e data r e t r i e v a l f i l e . RTECS ( R e g i s t r y of Toxic E f f e c t s of Chemical Substances) i s a product of the N a t i o n a l I n s t i t u t e f o r Occupational Safety and H e a l t h (NIOSH) and i s a v a i l a b l e as an annual p u b l i c a t i o n (2). NLM o b t a i n s a computer-readable f i l e from NIOSH who maintains r e s p o n s i b i l i t y f o r f i l e contents. The present (1977) v e r s i o n of RTECS contains t o x i c i t y data f o r approximatel provides a d e s c r i p t i o The RTECS f i e l d s i d e n t i f i e d w i t h a search a b b r e v i a t i o n are d i r e c t l y searchable (up t o 36 contiguous c h a r a c t e r s ) . The i n f o r m a t i o n i n the T o x i c i t y Data Index S t r i n g s (IX) i s l i n k e d t o prevent f a l s e a s s o c i a t i o n s and i s s t r u c t u r e d as f o l l o w s : ROUTE;SPECIES;ORDER;STUDY TYPE;VALUE; TOXIC EFFECTS ( i f any);SPECIFIC

EFFECT/ORGAN SYSTEM AFFECTED

An example of t h i s c a t e g o r i z a t i o n would be: ORAL;RAT ;RODENTS ;LD50; 54 MG/KG;TOXIC EFFECTS ;CARCINOGENIC Each term or phrase between semicolons i s d i r e c t l y s e a r c h able and can be ANDed together. For example, (IX)

ORAL AND RAT AND LD50 AND CARCINOGENIC

The i n f o r m a t i o n contained i n the f i e l d s t h a t are a s t e r i s k e d i n Table 1 are searchable as f r e e t e x t terms; t h a t i s , unique words which can be ANDed or ORed together. The (CT) f i e l d i d e n t i f i e r i s used when searching the Toxic Data Source f i e l d and (TW) i s used when searching f o r i n f o r m a t i o n i n the remaining "free text" f i e l d s . CHEMLINE-TOXLINE Search The previous d i s c u s s i o n s presented an overview of three of the N a t i o n a l L i b r a r y of Medicine's o n - l i n e f i l e s ; next w i l l be an examination of the v a r i o u s ways i n which CHEMLINE can be u t i l i z e d i n enhancing the chemical a c c e s s i b i l i t y t o t o x i c o l o g i c a l i n f o r m a t i o n i n another f i l e . Using the p e s t i c i d e , Leptophos, as an example, F i g u r e s 4 and 5 show how t h i s i s done.


64

RETRIEVAL

NLM

O F MEDICINAL

BIOS IS

MEDLARS

SECT 1-5 62-64 59,60

PROFILES

ASHP

EPA PESTICIDES ABSTRACTS

BIOL. ABSTR. & BioRESEARCH INDEX

CHEMICAL ABSTRACTS

CHEMICAL

ORNL

INT. PHARM. ABSTRACTS

SPECIAL FILES

PROFILES

[ TOXBIB

J

(

CBAC

Figure 3.

INFORMATION

)

(

HEEP

( V

Λ J

PESTAB Λ ΗΛΡΑΒ J

Sources of TOXLINE

(

ΓΡΤ

V

3(

EMIC

)

ET'IC

)

bibliographic files

SS 1

/c?

USER: "FILE CHEMLINE PROG: YOU ARE NOW CONNECTED TO THE CHEMLINE F I L E . SS 1 / C ? USER: (SY) LEPTOPHOS PROS: SS ( 1 ) PST6 ( 1 ) SS 2 / C ? USER: "PRT RN,SY,LO PROG:

RN SY SY SY SY SY SY SY SY SY LO

Figure 4.

CHEMLINE search

synonym

o LO L

-

21609-90-5

- VCS 506 - VELSICOL VCS 506 - VELSICOL 505 - PHOSVEL - K62-105 - LEPTOPHOS - FOSVEL - NK 711 - ABAR - TOXLINE - MEDLARS - EPATSCALIST


5.

SPANN E T A L .

CHEMLINE

65

Searching the synonym f i e l d of CHEMLINE f o r Leptophos r e t r i e v e s one r e c o r d . PRINTing the RN, SY and LO f i e l d s provides the CAS R e g i s t r y Number, other synonyms which are known for the compound and the source of a d d i t i o n a l i n f o r m a t i o n . As seen i n Figure 5, a s w i t c h i s made t o the TOXLINE f i l e and the g e n e r i c name f o r the p e s t i c i d e i s entered. In t h i s case, 160 records are r e t r i e v e d i n which the term Leptophos occurs i n the t i t l e , a b s t r a c t or keyword f i e l d s . In Search Statement 2, the CAS R e g i s t r y Number and a few of the synonymous terms from CHEMLINE are ORed (combined) as search parameters. This r e s u l t s i n the r e t r i e v a l of 257 records cont a i n i n g the d e s i r e d chemical substance. Thus, the CHEMLINE i n f o r m a t i o n e f f e c t i v e l y i n c r e a s e s the TOXLINE r e t r i e v a l by 97 r e c o r d s . Search Statement 3 i s designed t o s e l e c t those c i t a t i o n s published i n 1977 i n which both Leptophos and some v a r i a t i o n of the term nerve The t i t l e and source o c r i t e r i a are then PRINTed. N a t u r a l l y , Statement 3 i s not intended t o provide a comprehensive search f o r the b i o l o g i c a l concept s i n c e other terms such as CNS or n e u r o t o x i c i t y are not i n c l u d e d . However, the l i s t i n g of the output from t h i s search could a s s i s t the user i n s e l e c t i n g b i o m e d i c a l terms r e l e v a n t t o the scope of the search. CHEMLINE Substructure Searches The f o l l o w i n g s e c t i o n s are provided t o demonstrate s e v e r a l approaches t o s u b s t r u c t u r e searching i n CHEMLINE (3-6). F i g u r e 6 shows the s t r u c t u r a l diagram of Leptophos and the c h a r a c t e r i s t i c s of a substructure search query based on the chemical s t r u c t u r e of this insecticide. A l o g i c a l approach t o t h i s query would be t o search CHEMLINE for the given common name, PRINT the systematic chemical name and s e l e c t name fragments that correspond to the d e s i r e d subs t r u c t u r e . This i s i l l u s t r a t e d i n the f i r s t p o r t i o n of F i g u r e 7. The name fragment PH0SPH0N0THI0IC i s s e l e c t e d s i n c e i t i s the fundamental or one of the most s i g n i f i c a n t f e a t u r e s of the chemical substance. PHENYL, as a name fragment, would e x p l i c i t l y exclude any s u b s t i t u t i o n s on the r i n g (such as c h l o r o p h e n y l ) . This i s a requirment f o r Ring A as shown i n F i g u r e 6. The coord i n a t i o n of the two name fragments r e s u l t s i n the r e t r i e v a l of 34 r e c o r d s . The formula fragments CL and BR along w i t h the name fragment 4 r e q u i r e s t h a t a t l e a s t one c h l o r i n e and one bromine atom be present and the 4 - p o s i t i o n of a r i n g ( o r chain) be s u b s t i t u t e d . These requirements r e t r i e v e 294 records from CHEMLINE. The i n t e r s e c t i o n of those records r e t r i e v e d i n Search Statements 2 and 3 should s a t i s f y the c r i t e r i a e s t a b l i s h e d f o r the s u b s t r u c t u r e search (6 r e c o r d s ) .


66

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

SS 2 /C? USER: " F I L E TOXLINE PROG: YOU ARE NOW CONNECTED TO THE TOXLINE F I L E . SS 1 / C ? USER: LEPTOPHOS PROG: SS ( 1 ) PSTG ( 1 6 0 ) SS 2 / C ? USER: 2 1 6 0 9 - 9 0 - 5 OR PHOSVEL OR FOSVEL OR LEPTOPHOS PROG: SS (2) PSTG ( 2 5 7 ) SS 3 / C ? USER: 2 AND ALL NERV: AND 77 (YP)

USER: "PRT 1 T I , SO PROG: 1 TI

Figure 5.

TOXLINE

- NEUROTOXICITY OF ORGANOPHOSPHORUS INSECTICIDES LEPTOPHOS AND EPN.

so - J ENVIRON SCI HEALTH ( Β ) ; VOL 12, iss

search

1 9 7 7 , P269-87

ο-ί-b* OCH3 LEPTOPHOS

SUBSTRUCTURE

CHARACTERISTICS: 1.

No SUBSTITUTION ON RING A

2.

ANY ESTER

3.

ONE OR MORE CHLORINE ATOMS ANYWHERE ON RING Β IN COMBINATION WITH A BROMINE ATOM ON THE ^-POSITION

Figure 6.

Characteristics of leptophos-related substructure search


5.

SPANN E T A L .

67

CHEMLINE

USER: (SY) LEPTOPHOS PROG SS (1) PSTG (1) SS 2 /C? USER: "PRT N l PROG: 1 NI

- PHOSPHONOTHIOIC ACID, PHENYL-, 0-( I-BROMO-2,5-DICHLOROPHENYL) 0-METHYL ESTER (9CI) Î

SS 2 /C? USER: (NF) PHOSPHONOTHIOIC AND PHENYL PROG: SS (2) PSTG (34) SS 3 /C? USER: (FF) CL AND BR AND 4 PROG: SS (3) PSTG (294) SS 4 /C? USER: 2 AND 3 PROG: SS (4) PSTG (6) SS 5 /C? USER: "PRT Nl

Figure 7.

ÏLWimt

CHEMLINE

SIARCH_ÔMR£y^ PiLNT .AJMEY. (SI)

SI

CHEMICAL SUBST. PRIME NAME

(Nl)

Nl

CHEMICAL NAME FRAGMENTS

SOURCE

IDENTIFICATION

(NF)

--

CHEMICAL DEFINITION

•

CD

CAS REGISTRY NUMBER

(RN)

RN

MOLECULAR FORMULA

(MF)

MF

MOLECULAR FORMULA FRAGMENTS

(FF)

-

MOLECULAR WEIGHT

-

nw

WISWESSER LINE NOTATION

(WL)

WL

SYNONYMS

(SY)

SY

TOXICITY DATA INDEX STRINGS

(IX)

TDKW

so

TOXIC DATA SOURCE AQUATIC TOXICITY RATING

*

TOXICOLOGY & CANCER REVIEW

•

STANDARDS a REGULATIONS

*

NIOSH CRITERIA DOCUMENTS

•

NC

•

ST

STATUS

AQ TC Ç

R

TEXT WORDS

(TW)

--

CITATIONS TERMS

(CT)

so

• #

*

substructure search

SEARCHABLE AS TEXT WORDS SEARCHABLE AS CITATIONS TERMS

Table L

RTECS—Unit

Record Description


68

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

The systematic chemical names are then PRINTed to compare the search r e s u l t s w i t h the d e s i r e d s u b s t r u c t u r e query. The s t r u c t u r a l diagrams f o r the substances r e t r i e v e d are i l l u s t r a t e d i n F i g u r e s 8 and 9. CHEMLINE-RTECS Substructure Search The next example of a CHEMLINE s u b s t r u c t u r e search i s f o r c h l o r i n a t e d d i b e n z o d i o x i n s . As seen i n Figure 10, t h i s subs t r u c t u r e query i s d e r i v e d from the h e r b i c i d e contaminant D i o x i n . The search can be approached without u s i n g any chemical nomenclature through use of the Formula Fragment (FF) and r i n g informat i o n f i e l d s (Ring Elemental (RE) a n a l y s i s and Component Line (CL) formula). The search r e t r i e v e s f o r t y r e c o r d s ; a few of the chemical names are PRINTed t o review the output Since systematic chemica mula fragments are a l s CHEMLINE nomenclatural output to perform a s u b s t r u c t u r e search that i s c o r r e l a t e d w i t h a b i o l o g i c a l concept i n RTECS. T h i s would o b v i a t e the need t o c a r r y the chemical i d e n t i f i e r s f o r the f o r t y CHEMLINE records i n t o the RTECS f i l e . A f t e r accessing the RTECS f i l e (Figure 11), the name f r a g ments common t o each name l i s t e d i n the CHEMLINE search (DIBENZO and DIOXIN) and the formula fragment CL are used t o r e t r i e v e s i x records. The RTECS T o x i c i t y Data f i e l d s (IX) are then searched f o r the terms ORAL, RAT and TERATOGENIC and the records c o n t a i n ing these terms i n the index s t r i n g are i n t e r s e c t e d w i t h Search Statement 1 t o r e t r i e v e two records (Search Statement 2 ) . F i n a l l y , p a r t of the f i r s t RTECS record i s l i s t e d t o o b t a i n the t o x i c data f o r the c h l o r i n a t e d d i b e n z o d i o x i n . Summary This paper has d e s c r i b e d s e v e r a l approaches t o the u t i l i z a t i o n of an o n - l i n e chemical d i c t i o n a r y f i l e i n l i n k i n g chemical s t r u c t u r e s t o b i o l o g i c a l i n f o r m a t i o n . Through use of nomenclat u r a l , molecular formula and r i n g screens, the c a p a b i l i t y e x i s t s i n CHEMLINE t o e f f e c t i v e l y i d e n t i f y chemical substances w i t h s p e c i f i c s t r u c t u r a l c h a r a c t e r i s t i c s . Since CHEMLINE i s an i n t e g r a l p a r t of the N a t i o n a l L i b r a r y of Medicine's On-Line S e r v i c e s , i t i s then p o s s i b l e t o c o r r e l a t e the s u b s t r u c t u r a l data w i t h b i o l o g i c a l i n f o r m a t i o n and data contained i n b i b l i o g r a p h i c and/or data f i l e s e x i s t i n g w i t h i n the same computer environment.


SPANN

E T AL.

1 NI

CHEMLINE

- PHOSPHONOTHIOIC ACID, PHENYL-,

0-(4-BROMO-2,5-DICHLOROPHENYL)

ESTER, POTASIUM SALT (9CI) Cl

2 NI

- PHOSPHONOTHIOIC ACID, PHENYL-, 0-(4-BROMO-2,5-DICHLOROPHENYL) 0-METHYL ESTER, MIXT, WITH 1,1'-(2,2,2-TRICHLOROETHYLIDENE)BIS( I-CHLOROBENZENE) (9CI) £

NI

- PHOSPHONOTHIOIC ACID, PHENYL-, 0-(4-BROMO-2-CHLOROPHENYL) 0-METHYL ESTER (9CI)

Figure 8.

Substructure search results


RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

PHOSPHONOTHIOIC ACID, PHENYL-, 0-(4-BROMO-2,5-DICHLOROPHENYL) 0-METHYL ESTER (9C1)

PHOSPHONOTHIOIC ACID, PHENYL-, 0-(4-BROMO-2,5-DICHLOROPHENYL) 0-ETHYL ESTER (9CI)

PHOSPHONOTHIOIC ACID, PHENYL-, 0-(4-BROMO-2,6-DICHLOROPHENYL) 0-METHYL ESTER (9CI)

Figure 9.

Substructure search results


5.

SPANN

71

CHEMLINE

E T AL.

DIOXIN

Cl

CHLORINATED DIBENZODIOXINS USER: (RE) C402-C6-C6 AND 0C20C2 (CL) AND CL (FF) PROG: SS iS) PSTG (40) SS 2 /C? USER: "PRT 4 NI PROG: 1 NI - DIBENZ0(B,E)(1,4)DI0XIN,

TETRACHLORO- (9CI)

2 NI - DIBENZ0(B,E)(1,4)DI0XIN,

1,2.3.7.8-PENTACHL0R0- (9CI)

3 NI - DIBENZ0(B,E)(1,4)DI0XIN,

1,2,3,4,6,7,8-HEPTACHLORO- (9CI)

4 NI - DIBENZ0(B,E)(1,4)DICXIN,

Figure 10.

HEXACHLORO- (9CI)

Chlorinated dibenzodioxins search

SS 2 /C?

USER: "FILE RTECS PROG: YOU

ARE NOW CONNECTED TO THE RTECS FILE.

SS 1 /C? USER: (NF) DIBENZO AND DIOXIN AND CL (FF) PROG: SS (1) PSTG (6) SS 2 /C? USER: (IX) ORAL AND RAT AND TERATOGENIC AND 1 PROG: SS (2) PSTG (2) SS 3 IV. USER: "PRT 1 TOXDATA PROG: SI

- NI0SH/HP32000

Nl

- DIBENZO-P-DIOXIN, HEXACHLORO-

RN

- 34465-46-8

SO

- ADCSAJ ADVANCES IN CHEMISTRY SERIES. 120,55,73

TDKW- ORAL;RAT;RODENTS;LDLo; 100 MG/KG

ORAL; RAT;RODENTS;TDLo;100

UG/KG/(6-15D

PREG)

; TOXIC EFFECTS

TERATOGENIC

Figure 11.

RTECS search



72

ABSTRACT The National Library of Medicine's (NLM) on-line chemical dictionary f i l e (CHEMLINE) is primarily used to enhance the retrieval of bibliographic information associated with chemical substances. This discussion demonstrates the utility of CHEMLINE as a mechanism to link chemical substructures to biological data. Search techniques are developed to integrate classes of structur ally related chemicals with toxicity data and information con tained in on-line retrieval files such as the Registry of Toxic Effects of Chemical Substances (RTECS) and TOXLINE. Acknowledgements The authors gratefull by Dr. Henry M. Kissman and Mrs. Joan H. Cepko of the Toxicology Information Program. Literature Cited 1.

2.

3.

4.

5.

6.

S c h u l t h e i s z , R. J., Kannan, K. L. and Walker, D. F., "Design and Implementation of an O n - l i n e Chemical D i c t i o n a r y (CHEMLINE)", J. Am. Soc. I n f . Sci., Accepted f o r p u b l i c a t i o n in Vol. 29 (1978). For s a l e by the Superintendent of Documents, U . S . Government Printing O f f i c e , Washington, D . C . 20402 GPO Stock No. 017-033-00271-1. F i s a n i c k , W., Mitchell, L. D., S c o t t , J.A., and Vanderstouw, G. G., "Substructure Searching of ComputerReadable Chemical A b s t r a c t s S e r v i c e N i n t h Collective Index Nomenclature Files", J. Chem. I n f . Comput. Sci., 15 (2) 73-84 (1975). Dunn, R. G., F i s a n i c k , W., and Zamora, Α . , " A Chemical Substructure Search System Based on Chemical A b s t r a c t s Index Nomenclature", J. Chem. I n f . Computer. Sci., 17 ( 4 ) , 212-219 (1977). "Substructure Searching of Computer-Readable CAS 9CI Chemical Nomenclature Files (Based on Nomenclature in the N i n t h Collective Index of Chemical A b s t r a c t s ) (1972-1976)", Chemical A b s t r a c t s S e r v i c e , Columbus, Ohio, Aug. 1974, 128 p p . , ISBN 8412-0204-4, LCN 74-14778. V a s t a , B . M . and Spann, M. L., "Chemical Searching C a p a b i l i t i e s of CHEMLINE", presented at the 172nd N a t i o n a l Meeting of the American Chemical S o c i e t y , August, 1976.



6

C h e m i c a l a n d Biological Data—an Integrated

On-Line

Approach

Ε. E. TOWNSLEY and W. A. WARR Data Services Group, ICI Pharmaceuticals Division, PO Box 25, Alderley Park (Mereside), Macclesfield, Cheshire, SK10 4TG, England For some years CROSSBO technolog (1-5) ha bee d b ICI Pharmaceuticals Division database. A computerised system f o r biological c o n t r o l and f o r the storage and retrieval of biological data has a l s o been i n o p e r a t i o n for many y e a r s . However, the biological database was h e l d separately from the chemical database and there were some i n t e r f a c e problems. The advantage of the new i n t e g r a t e d chemical and biological system over the previous two separate systems i s the ease of access ( i n c l u d i n g o n - l i n e access) to all i n f o r m a t i o n , both biological and c h e m i c a l . In the past it was not q u i t e so easy, for example, to f o l l o w a biological search w i t h a search of the chemical database, or v i c e v e r s a . When the Division installed a Burroughs 6700 computer i t was necessary to redesign the databases and convert all programs which had earlier run on a Burroughs 4700 machine. I t was decided that development, improvement and conversion should be undertaken at the same time, and an i n t e g r a t e d o n - l i n e database is now in use. This ever-expanding database at present holds 190,000 chemical compounds from f i v e ICI divisions: data on 267 distinct biological t e s t s : and over 1,000,000 biological test results from Pharmaceuticals Division. Linked to the computer are three Burroughs t e r m i n a l computers and eight v i s u a l d i s p l a y units. The system's uses are chemical and biological data registration and retrieval, and biological c o n t r o l . The b i o l o g i c a l c o n t r o l system manages the processing of samples submitted for b i o l o g i c a l assay. Data Services Group takes a compound from the chemist, weighs appropriate samples for a l l the necessary b i o l o g i c a l t e s t s and c o l l e c t s t e s t r e s u l t s from the b i o l o g i s t s . The register allocate example, 01 where

f i r s t step i n the b i o l o g i c a l c o n t r o l sequence i s to the s t r u c t u r e of a compound, and i f i t i s n o v e l , to i t the next a v a i l a b l e s i x - f i g u r e M number, say, f o r ΜΙΟΟ,ΟΟΟ. This f i r s t sample i s r e f e r r e d to as M100000/ 01 i s c a l l e d a stroke number. I f the compound i s not 0-8412-0465-9/78/47-084-073$05.00 Published 1978 American Chemical Society


74

RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

n o v e l , the sample w i l l be a l l o c a t e d the next a v a i l a b l e s t r o k e number t o the p r e v i o u s l y a l l o c a t e d M number, e.g. M80996/09. The next step i s to r e g i s t e r the property data (see F i g u r e 1) of the given sample and t o a l l o c a t e a sample storage number, which i s the s h e l f storage number f o r the stock b o t t l e . The handling codes recorded on the property f i l e are twol e t t e r codes i n d i c a t i n g whether any unusual h a n d l i n g precautions need to be observed (e.g. TX f o r " t o x i c " , KC f o r "keep c o l d " ) . I n t e n t i o n - t o - t e s t data i s held on a dynamic f i l e whose record layout i s shown i n F i g u r e 2. For each sample/test combination an i n t e n t i o n - t o - t e s t record i s created and used t o manage the progress of the sample from the chemist's i n i t i a l submission through to the c o l l e c t i o n of t e s t r e s u l t s from the b i o l o g i s t , a f t e r whic removed. Most of the data i n F i g u r e 2 i s s e l f - e x p l a n a t o r y . A t e s t i s represented by two or three characters (e.g. AB o r AB3 f o r a n t i b a c t e r i a l t e s t s ) f o r e x t e r n a l use and a number f o r machine use. The sample property r e c o r d must be created before an i n t e n t i o n - t o - t e s t record can be made because data needed f o r the l a t t e r record i s a b s t r a c t e d from the former (compare F i g u r e s 1 and 2 ) . The a l l o c a t i o n of a p r i o r i t y i s important because i t ensures that compounds are progressed through the system i n such a way that important p r o j e c t compounds are always s e l e c t e d i n preference to non-project compounds where there are more t e s t submissions than a b i o l o g i s t can cope w i t h at any one time. Four p r i o r i t i e s are used. Compounds made by the chemist f o r t e s t i n g i n connection w i t h h i s own p r o j e c t are given p r i o r i t y 1 (the h i g h e s t ) . The chemist may e l e c t to send the same compound f o r t e s t s unconnected w i t h h i s p r o j e c t and the i n t e n t i o n - t o - t e s t records f o r these w i l l have p r i o r i t y 2. Compounds the chemist has s e l e c t e d by means of a s u b s t r u c t u r e search around a " l e a d " compound, w i l l be t e s t e d on p r i o r i t y 3 and screening compounds randomly s e l e c t e d by Data Services Group are tested under p r i o r i t y 4 (the l o w e s t ) . The date of submission i s recorded so t h a t , f o r a given p r i o r i t y , records created e a r l i e r are processed before those created l a t e r . The s t a t u s of a sample i s described i n F i g u r e 3. When an i n t e n t i o n - t o - t e s t record i s created i t has the s t a t u s value s e t to 1. This value i s updated stepwise from 1 to 6 as the sample progresses through the weighing and documentation process, followed by submission to the b i o l o g i s t and f i n a l l y reaches the r e s u l t c o l l e c t i o n and n o t i f i c a t i o n stage. When the t e s t c o n t r o l f u n c t i o n f o r the sample i s complete, the i n t e n t i o n record i s removed from the f i l e .


TOWNSLEY

A N D WARR

Chemical

and Biological

Data

M number/stroke number Key: Index s e q u e n t i a l Company r e g i s t r y number S u f f i x p o i n t e r (e.g. 0201 o r 0001 i n Figure 7) Sample a v a i l a b l e / n o t a v a i l a b l e / r e s e r v e d Sample storage number Chemist's i n i t i a l s and notebook reference S a l t data S o l u b i l i t y codes Handling codes Figure 1.

Property file (standard data set)

M number Stroke number Key Test number Priority Key Date 0 Status Test l e t t e r s Sample storage number Chemist's i n i t i a l s and s e c t i o n Handling codes Figure 2.

Status 1 2 3 4 5 6

Index s e q u e n t i a l Index s e q u e n t i a l

Intention-to-test file (standard data set)

Meaning Sample f o r t h i s t e s t unweighed and undocumented. Sample f o r t h i s t e s t weighed but undocumented. Sample f o r t h i s t e s t weighed and documented. Sample f o r t h i s t e s t submitted t o b i o l o g i s t . Test r e s u l t has been received from b i o l o g i s t . Test r e s u l t n o t i f i e d t o s e c t i o n leader and i n t e n t i o n - t o - t e s t record ready f o r removal. Figure 3.

Status of sample on intention-to-test file


76

RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

From the i n t e n t i o n - t o - t e s t f i l e i s generated a m a t r i x report which d e t a i l s the c u r r e n t s t a t e of sample p r o g r e s s i o n against t e s t . I t r e p o r t s the numbers of samples f o r each t e s t at each management l e v e l , up t o and i n c l u d i n g s t a t u s f o u r , subdivided by p r i o r i t y . Enquiry r o u t i n e s give more d e t a i l e d i n f o r m a t i o n at the s p e c i f i c sample l e v e l . B i o l o g i c a l t e s t s are of two types - c u r r e n t and screen. A current t e s t i s one which r e q u i r e s p r o j e c t compounds to pass as q u i c k l y as p o s s i b l e from chemist to b i o l o g i s t , and such compounds must be weighed and documented immediately a f t e r i n p u t . U s u a l l y there i s only a small number of samples to be processed each week f o r one of these t e s t s . A screen t e s t i s one which r e q u i r e s a r e g u l a r supply of a f i x e d number of sample made up of chemist's p r o j e c submitted compounds, and the balance i s provided from p r i o r i t y 3 and 4 submissions. To c a t e r f o r the two types of t e s t there are two ways i n which samples are s e l e c t e d f o r progress to weighed and documented s t a t u s and then t o submission t o the b i o l o g i s t . The f i r s t produces weighing i n s t r u c t i o n s , b o t t l e l a b e l s and t e s t documentation f o r samples s e l e c t e d by M number e s s e n t i a l l y current p r o j e c t compounds. U s u a l l y , a l l the t e s t samples from one stock b o t t l e are weighed at the same time, i r r e s p e c t i v e of t e s t type, so as to minimise b o t t l e h a n d l i n g . However, documentation f o r screen t e s t s i s produced a t a l a t e r stage. Thus the s t a t u s of c u r r e n t t e s t records i s updated to 3 and that of screen t e s t records t o 2, by t h i s "weighing r o u t i n e " . The second weighing r o u t i n e , which i s geared to screen t e s t requirements, s e l e c t s s t a t u s 1 and 2 records by t e s t . F o r each t e s t the samples are s e l e c t e d i n p r i o r i t y o r d e r , and w i t h i n p r i o r i t y , by date. Again, weighing i n s t r u c t i o n s and b o t t l e l a b e l s (where needed) and t e s t documentation are produced together w i t h two d i f f e r e n t working documents which d e t a i l sample numbers and r e l a t e d d a t a , one f o r the b i o l o g i s t ' s use and one to be used w i t h i n Data Services Group. A t e s t d e t a i l s f i l e i s maintained which holds constant data p e r t a i n i n g to each t e s t , as shown i n F i g u r e 4. When t e s t r e s u l t data i s returned i t i s v a l i d a t e d against the d e t a i l s h e l d on t h i s f i l e . At the same time as an i n t e n t i o n - t o - t e s t r e c o r d i s c r e a t e d , a " r e s u l t " entry i s created on the summary r e s u l t f i l e (see F i g u r e 5) w i t h the r e s u l t area l e f t blank. As the sample progresses to submitted s t a t u s the blanks are r e p l a c e d by "S". F i n a l l y t h i s area w i l l be occupied by the b i o l o g i s t ' s summary


6.

TOWNSLEY

A N D WARR

Chemical

and

Biological

Data

77

t e s t r e s u l t s (e.g. A f o r a c t i v e o r I f o r i n t e r e s t i n g ) f o r up t o four s u b d i v i s i o n s of h i s t e s t . A screen f i l e (Figure 6) i s maintained a t the compound l e v e l as opposed to the sample l e v e l . Each compound i s a l l o c a t e d a sample a v a i l a b i l i t y b i t and one b i t per t e s t which i s used to i n d i c a t e whether or not any sample of the compound has been submitted f o r that t e s t . This f i l e i s maintained because the summary t e s t r e s u l t f i l e i s very l a r g e and before i t i s searched for t e s t r e s u l t i n f o r m a t i o n the screen f i l e i s q u i c k l y scanned so as to reduce the number of accesses that need be made t o the larger f i l e . To r e t u r n t o the beginning of the r e g i s t r a t i o n process, a chemical s t r u c t u r e i s input as a molecular formula and a Wiswesser L i n e N o t a t i o n (WLN) (6-7) The f i r s t o n - l i n e r o u t i n e generates a molecular formul agrees w i t h the input c a r r i e d out a g a i n s t the WLNs already on f i l e . I f the compound i s novel i t s WLN i s r e g i s t e r e d and a fragment screen i s generated from the n o t a t i o n and stored (separate from the WLN) on a d i r e c t data s e t w i t h the company r e g i s t r y number (CR number) as key. There are 148 designated fragments, the presence of which are i n d i c a t e d by b i t s s e t to one r a t h e r than zero. Each fragment i s r e l a t e d t o a s i n g l e WLN c h a r a c t e r , o r a group o f them, o r t o a f e a t u r e of a WLN c i t e d r i n g . Ring s u b s t i t u e n t fragments are d i f f e r e n t i a t e d from s i m i l a r fragments in a chain. The way WLN data i s held i s shown i n Figure 7. I t can be seen from t h i s that a CR number represents a WLN, or a twodimensional s t r u c t u r e whereas a d i v i s i o n a l reference number (e.g. an M number) represents a compound. Sometimes i n f o r m a t i o n such as stereochemistry, has to be s u f f i x e d t o the WLN proper to give a f u l l r e p r e s e n t a t i o n of a compound. I n the many cases where there i s no s u f f i x , CR number and d i v i s i o n a l reference number are e q u i v a l e n t . The combination of WLN-last-16-inverted plus WLN length was chosen as key because f o r a very high p r o p o r t i o n of the records on the f i l e t h i s key i s unique. The mean length of n o t a t i o n s r e g i s t e r e d on the database i s 21 c h a r a c t e r s . The main WLN f i l e holds up t o 34 characters o f n o t a t i o n . A s m a l l number of WLNs (about 7%) have t o overflow onto a s u b s i d i a r y f i l e . T h i r t y - f o u r characters of WLN are held on the main f i l e , because that number reduces wasted space on the main and overflow f i l e s t o a minimum. The database i s designed so t h a t i t may be accessed by WLN, d i v i s i o n a l reference number, CR number or molecular formula.


78

RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

Test number Key 1 - d i r e c t Test l e t t e r s Key 2 - index s e q u e n t i a l F l a g - i s t e s t current o r screen? Biologist's i n i t i a l s Weight of sample r e q u i r e d f o r t e s t Stationery Test r e s u l Figure 4.

Test detaih file (direct data set)

Key - index s e q u e n t i a l M number/stroke number/record number Test l e t t e r s - p r i o r i t y ^) occurs once f o r each date t e s t on t h i s sample chemist summary r e s u l t

Figure 5.

Summary result file (standard data set)

M number Key - d i r e c t F l a g - sample a v a i l a b l e / n o t a v a i l a b l e Tested/untested b i t s f o r 480 p o s s i b l e t e s t s Figure 6.

Screen file (direct data set)


6.

TOWNSLEY A N D WARR

Chemical

and

Biological

Data

79

There i s a separate data s e t f o r compounds which cannot be coded i n t o WLN (e.g. seed e x t r a c t s and r e a c t i o n products of unknown s t r u c t u r e ) and these are s t o r e d on a random data s e t w i t h d i v i s i o n a l r e f e r e n c e number as key. The CROSSBOW chemical search system (1-5) i s a m u l t i l e v e l one. Search of the b i t screens i s f i r s t c a r r i e d out and t h i s q u i c k l y and cheaply reduces the f i l e to ten per cent (or l e s s ) of i t s o r i g i n a l s i z e . There w i l l almost c e r t a i n l y be many f a l s e drops but most of these can u s u a l l y be removed by s t r i n g search of the WLNs and/or molecular formulae and/or reference numbers. S t r i n g searching i s slower and more expensive than b i t search. Connection t a b l e generation and atom-by-atom search of the connection t a b l e s (the t h i r d search l e v e l ) are s t i l l slower and even more expensive, but the atom-by-atom search program i s a very powerful t o o l which i d i about 80% f a l l searches. The CROSSBO any search are f i n a l l y inpu display program. B i t and s t r i n g searching i s an i n t e r a c t i v e process. The paramaters are input on a VDU, the h i t count i s d i s p l a y e d o n - l i n e and, i f necessary, the search parameters can be modified and the search repeated. At the end of a s e s s i o n , search h i t f i l e s are merged as r e q u i r e d and connection t a b l e generation, atom-byatom search and s t r u c t u r e d i s p l a y are run batchwise. An average b i t and s t r i n g search takes about f i v e minutes (searching 190,000 compounds). The CROSSBOW connection t a b l e generation program handles 1500-2000 compounds a minute, and the atom-by-atom program searches 600 compounds a minute. Well over 90% of n o t a t i o n s are amenable to connection t a b l e generation and over 90% of the compounds on the database can be s t r u c t u r a l l y d i s p l a y e d d i r e c t l y from the connection t a b l e . (The remainder are held on the database i n a f i l e of " d i f f i c u l t " structures.) A b i t search i s s p e c i f i e d u s i n g combinations of AND, OR and NOT l o g i c , and a l s o nested AND-within-OR l o g i c , as r e q u i r e d . S t r i n g search allows the same l o g i c a l o p e r a t o r s , but i n a d d i t i o n there are ignore and followed-by f a c i l i t i e s and other s p e c i a l i s e d syntax t o s i m p l i f y the s p e c i f i c a t i o n of s e v e r a l alternative strings. F i g u r e 8 shows a t y p i c a l b i t and s t r i n g search. I n the b i t s search, fragments 50 and 137 must both be present. Neither fragment 134 nor fragment 135 may be present, and nor may be fragment 136. The OR clause s t a t e s that i f the combination of ( e i t h e r 17 or 18) and ( e i t h e r 71 or 72) i s not present, then, a l t e r n a t i v e l y the combination of ( e i t h e r 67 o r 68) and ( e i t h e r 28 or 29) must be present. The Wiswesser s t r i n g s which could represent t h i s s u b s t r u c t u r e a r e :


RETRIEVAL

80

O F MEDICINAL

CHEMICAL

INFORMATION

M number nnnnnn represents (WLN+suffix) Λ CR number NNNNNN represents WLN k

Λ

,

T66 BOT&J C3Y1&3Y1&3Y1&1 C l G l HQ I I J l &&DL FORM Overflow record NNNNNN/0001

Overflow record NNNNNN/0201

WLN F i l e Record (Direct Data Set) NNNNNN

CR number.

Key - d i r e c t .

1J I I QH 1G 1C 1 WLN l e n g t h . 0039 T66 B0T&J...G1 HQ

Up t o 34 characters of WLN.

C29H5002

M o l e c u l a r formula. + Various d i v i s i o n a l references and f l a g s . Figure 7.

Key - index random.

Storage of chemical structural information

X = 0CH or NHR (R^H) 3

Only the one r i n g present.

Hal Subst halogen One benzene r i n g Subst CO.O Subst CO Chain methyl Chain NH Ring systems

50 137 71 67 17 28 134

72 68 18 29 135 136

Parameters AND,BITS "50 137". Ν 0 Τ , /134 135 136/. 0R,/71 72/ /17 18/,/67 68/ /8 29/. %*#V##0M#.END 0R,WLN -l/l-"10VR ,"MVR , ff

M

Figure 8.

fl

Bit and string search


6.

TOWNSLEY AND WARR

Chemical

and

Biological

Data

81

Notations beginning "10VR" or n o t a t i o n s c o n t a i n i n g "MVR" or n o t a t i o n s c o n t a i n i n g xVO" or " xVM" 11

where χ = any character but space (represented by % i n Figure 8). The l a s t two s t r i n g s are combined together, the hash marks e n c l o s i n g c h a r a c t e r s which are a l t e r n a t i v e s . Atom-by-atom search i s r e s o r t e d to i f the s t r i n g search of WLNs and/or molecular formulae and/or reference numbers i s not s p e c i f i c enough. This w i l l occur i f , f o r example, the s u b s t r u c t u r e i s very branched or may be p a r t i a l l y o r wholly embedded i n a r i n g . An atom-by-atom search may a l s o be needed i f there i s a s p e c i f i c p o s i t i o n a l r e l a t i o n s h i p between two groups and t h a t r e l a t i o n s h i by WLN s u b - s t r i n g s . Th l a s t category. An atom-by-atom search t o f i x the s p a t i a l r e l a t i o n s h i p between the halogen and the carbonyl groups i s shown i n F i g u r e 9. E, F, G, I , D, T, U, 0, Q and M are the p o s s i b l e ways the nodes 1-7 could be represented i n the CROSSBOW connection t a b l e , a node being an atom and i t s surrounding bonds. NOTR stands f o r "not i n a r i n g " and RINGSA f o r a s i n g l e benzene r i n g (S) of the type shown i n the RINGSCREENA parameter. The RINGSCREENA parameter designates a benzene r i n g which has s u b s t i t u e n t s a t the 1 and 3 p o s i t i o n s and o p t i o n a l l y s u b s t i t u e n t s a t other p o s i t i o n s a l s o . The seventh node must not be t e r m i n a l (or the Q would represent OH o r the M, NH2) so the L before the QM i n d i c a t e s t h a t t h i s node must be l i n k i n g . The network parameter B65X, i n d i c a t e s that the seventh node i s attached t o the f i f t h . I f d e s i r e d , a b i o l o g i c a l search may be c a r r i e d out before or a f t e r the chemical search. This may perform the f u n c t i o n of reducing the t o t a l search output, or i t may be c a r r i e d out merely t o produce a l l the t e s t and sample data f o r each compound which i s a h i t on the chemical search. The operator MNOS i s used t o s p e c i f y the range of numbers to be searched and, i f r e q u i r e d , to l i m i t the number of h i t s . The operator OUT i s used t o s p e c i f y what sample and r e s u l t data i s to be p r i n t e d f o r each h i t . The operators EVER and NEVER ( s i m i l a r to OR and NOT) are used t o search the data on the property f i l e and the summary r e s u l t f i l e , u s i n g the screen f i l e , i f r e l e v a n t , as a p r e l i m i n a r y screen. Thus, f o r example, one may search f o r a l l the a c t i v e a n t h e l m i n t i c s Dr Robinson made and reserved i n 1977; or a l l the s o l i d s i n l a r g e s i z e b o t t l e s which are a c t i v e i n an a n t h e l m i n t i c t e s t but have not been t e s t e d i n any of the l i v e r f l u k e t e s t s .


82

RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

= NH or 0.

Parameters RINGSCREENASTCTCCC 1N0TR EFGI 2RINGSA Τ 3RINGSA DT 4RINGSA Τ 5N0TR TU 6N0TR 0 7N0TR LQM B65X Figure 9.

Atom-by-atom search

AA MNOS,1-20000=1000. EVER, (ACTIVE AB2 ) (AVAIL) . NEVER,(TESTED"AB3 ). OUT,BSLONG BRLONG ALL.END ,f

!!

n

Figure 10.

Biological search input

AA 13/03/78 *123456 MOLECULAR FORMULA C9H1003 WISWESSER NOTATION QYR&V01

SEARCH OUTPUT

HO--CH-//

I

L FORM

Ml31833

S

\.„/

C0-0-CH3

OTHER REF NUMBERS R12345 STROKE TEST NUMBER LETTERS 01

LFl

RESULT Ν

WEEK YEAR 48

Figure 11.

77

CHEM. WAW

SECT. PRIORITY 99

2

Search output


6.

TOWNSLEY AND WARR

Chemical and Biological Data

83

F i g u r e 10 shows a b i o l o g i c a l search f o r 1000 compounds i n the range Ml to M20000, these compounds being a c t i v e i n the AB2 t e s t , having samples a v a i l a b l e and having never been t e s t e d i n the AB3 t e s t . The OUT parameter s t a t e s that f u l l sample data must be output and f u l l r e s u l t d e t a i l s f o r every t e s t to which each compound has been submitted. The output from a chemical search i s 8" χ 5" c a r d s , each card bearing a s t r u c t u r e drawn from the connection t a b l e by the computer. An example (not r e l a t e d to the two searches above) i s shown i n F i g u r e 11. Planned Developments Wiswesser L i n e N o t a t i o n i s l i k e l y to be used at ICI Pharmaceuticals D i v i s i o n f o r many years to come, not l e a s t because of i t s use i n commercial database bought b I C I f o example the Index Chemicu We are u n l i k e l y to put t h i s vast database o n - l i n e , although we r e g u l a r l y manipulate i t and search i t u s i n g the CROSSBOW system. We do, however, have immediate plans to augment our o n - l i n e database by adding v a r i o u s sample property and t e s t r e s u l t f i l e s for two more ICI d i v i s i o n s , and by adding the Commercially A v a i l a b l e Organic Chemicals Index (CAOCI) being b u i l t up by a number of European companies. Acknowledgement We would l i k e to express our thanks to Paul Bowler (who was instrumental i n the design of the database) f o r h i s h e l p f u l reading and c r i t i c i s m of t h i s paper. Literature Cited 1.

Hyde, E., Matthews, F . W . , Thomson, L.H. and Wiswesser, W . J . - "Conversion of Wiswesser N o t a t i o n to a C o n n e c t i v i t y M a t r i x f o r Organic Compounds". J.Chem.Doc., V.7 ( 4 ) , p.200-204 (1967).

2.

Thomson, L.H., Hyde, E . and Matthews, F.W. - "Organic Search and D i s p l a y Using a C o n n e c t i v i t y M a t r i x Derived from the Wiswesser N o t a t i o n " . J.Chem.Doc., V.7 (4), p.204-207 (1967).

3.

Hyde, E . and Thomson, L.H. - " S t r u c t u r e D i s p l a y " . J.Chem.Doc., V . 8 , p.138-146, 1968.

4.

E a k i n , D . R . - "The ICI CROSSBOW System" and Ash, J.E. "Connection Tables and Their Role in a System" i n Ash, J . E . and Hyde, E. - "Chemical Information Systems". Horwood, 1975.



84

5·

E a k i n , D . R . , Hyde, E . and Palmer, G. - "The Use of Computers w i t h Chemical S t r u c t u r a l Information: ICI CROSSBOW System". P e s t i c i d e Sci. p.319-326, 1973.

6.

Smith, E . G . and Baker, P . A . - "The Wiswesser Line-Formula Chemical N o t a t i o n (WLN)". 3rd edition, CIMI, New J e r s e y , 1975.

7.

Baker, P . Α . , N i c h o l s , P.W.L. and Palmer, G. - "The Wiswesser Line-Formula N o t a t i o n " in "Chemical Information Systems" (loc.cit. at 4 ) .

8.

ICRS

® tapes c o n t a i n data from Current Abstracts of

Chemistry and Index Chemicus™ and both are a v a i l a b l e from the I n s t i t u t Chestnut S t r e e t RECEIVED August 29, 1978.


7

U s e of P r o p r i e t a r y B i o l o g i c a l a n d C h e m i c a l Merck

D a t a at

& Co., I n c .

I. M. R. EGGERS, W. B. GALL, F. A. CUTLER, JR., and H. D. BROWN Merck Sharp & Dohme Research Laboratories, Rahway, NJ 07065

L i k e other pharmaceutica Dohme Research Laboratorie proprietary biological and chemical data. Moreover, the scientific s t a f f at MSDRL is using these data in i n c r e a s i n g l y s o p h i s t i c a t e d ways in their search f o r products to improve human and animal h e a l t h . An o u t l i n e of the systems (shown in F i g u r e 1) will be h e l p ful f o r understanding t h i s usage. Biological Data Biodata Systems S t r u c t u r e . The beginning of the biological data system grew out of the n o t i o n that it would be helpful to have all the data on a given compound immediately a v a i l a b l e in a single listing. From the original hand-posted card file, a punched card record was developed, using the concept of one record f o r one test. I n 1968 a computer tape record was s t a r t e d . This computer-stored file now c o n s i s t s of 2.5 million records i n s e v e r a l hundred t e s t s representing 135,000 compounds. The b a s i c record as it appears f o r entry i n t o computer storage is o u t l i n e d in Figure 2. Each record answers the f o l l o w i n g q u e s t i o n s , e i t h e r by summarizing the experiment i t s e l f or by l i n k i n g or p o i n t i n g to other f i l e s . 1.

When was the t e s t done?

2.

What was being tested? The 1 1 - d i g i t L-number i s the l i n k to the Chemdata f i l e s which w i l l supply the chemical name and, i f d e s i r e d , the s t r u c t u r e .

3.

How was the t e s t performed? The 4 - d i g i t t e s t number p r o v i d e s the l i n k to the f i l e of d e s c r i p t i o n s of the s e v e r a l hundred t e s t s c u r r e n t l y i n computer s t o r a g e . V a r i a t i o n s on the b a s i c t e s t , such as d i f f e r e n t routes of a d m i n i s t r a t i o n 0-8412-0465-9/78/47-084-085$05.50 Published 1978 American Chemical Society In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

RETRIEVAL

86

O F MEDICINAL

CHEMICAL

INFORMATION

B i o l o g i c a l Data Biodata Systems S t r u c t u r e Records A c q u i s i t i o n Search S t r a t e g i e s and Examples Chemical Data Chemical S t r u c t u r e Information Systems (CSIS) Chemical Name Compound R e p o s i t o r Transaction F i l e Chemdata Enhancements Transaction File/Inventory Interaction Figure 1.

1-•6 7-•8 9 10-•11 12-•15 16-•40 41-43 44-•72 73 74 75-•80

L-Number S a l t Form Check L e t t e r Batch Test Number Test Name P r o t o c o l Code Test R e s u l t Response Code Dose Code Test Date Figure 2.

Outline

Must be present and numeric Must be present and numeric Must pass program which checks c o l s . 1-8 Must be numeric , i f present Must be present and numeric Free Form Must be present and numeric Free Form Must be present and numeric Must be numeric , i f present Must be present - month, day, year

Basic record—biological

data


7.

EGGERS E T A L .

Proprietary

Biological

and

Chemical

Data

87

of t e s t compound, o r dosing schedules, are defined by the 3 - d i g i t p r o t o c o l codes. At present, the number of v a r i a t i o n s i s i n excess of 2,400. D e f i n i t i o n s are on each Data Sheet, very abbreviated. 4.

Why do we run t h i s t e s t ? The r a t i o n a l e i s a part of the p r o t o c o l , which i s w r i t t e n i n d e t a i l , dated, and signed by the b i o l o g i s t i n charge of the t e s t , and i s f i l e d i n the Biodata Department.

5.

Where was the t e s t done and who d i d i t ? These items are noted on the Data Sheet, as w e l l as being i n c o r p o r a t e d i n the P r o t o c o l write-up.

There are 53 column c a l l e d " f r e form" whos content v a r i e s w i t h the t e s t e s t i s described Biologica (se F i g u r 3). When p r o p e r l y f i l l e d out, these forms c o n s t i t u t e the set of d i r e c t i o n s f o r the keypunch operators. These are made out by the Biodata Department i n cooperation w i t h the b i o l o g i s t r e s p o n s i b l e f o r the t e s t and the chemist r e s p o n s i b l e f o r s u p p l y i n g m a t e r i a l s to the t e s t . These " f r e e form" items make the system b a s i c a l l y open ended. Among the parameters entered i n t o these " f r e e form f i e l d s " are those l i s t e d i n F i g u r e 4. The dose u n i t s may be e x p l i c i t , abbreviated, or implied. O c c a s i o n a l l y , the number of parameters to be recorded i s so l a r g e that the a c t u a l record i s simply a s e r i e s of numbers. I f necessary, two or more records are generated. A l l of these records c o n s t i t u t e d a t a , as opposed t o documents, and t h i s d i f f e r e n c e i s most s i g n i f i c a n t . The P r o t o c o l f i l e i s a document f i l e . I n the planning stage i s a Term Index to the p r o t o c o l f i l e , c o n t a i n i n g i n a searchable form such items as s p e c i e s , r o u t e of a d m i n i s t r a t i o n , whether enzyme, v i r u s , yeast, o r other type of t e s t , date of f i r s t record i n computer storage, name of i n v e s t i g a t o r , dose u n i t s , and p o s s i b l y others. Various other f i l e s are maintained f o r such purposes as housekeeping, a n a l y s i s of a c t i v i t i e s , and cross reference (see Figure 5). A l s o , there are some manually-posted f i l e s of b i o l o g i c a l data which predate the computer system. That i s the s t r u c t u r e of the system. Data must be acquired to b u i l d i t i n the f i r s t p l a c e and then to maintain i t on a current b a s i s . Records A c q u i s i t i o n . Records may be sent to Biodata from any Merck l a b o r a t o r y s i t e o r o u t s i d e i n v e s t i g a t o r . These may be i n the form o f : (1) copies of l a b o r a t o r y notebook records (the most common), (2) punched cards from systems not compatible w i t h



BER

NUM.

TEST

BIOLOGIST AND/OR CHEMIST

nun iiiniiiiiniiiiiii

EXPANDED TEST TITLE (100 CHARACTER MAXIMUM)

LU

BIOLOGICAL D A T A SHEET

DATE COLUMNS

75-7* 77-7· 7*40

oro«$

SL C O . , I N C .

N l w j I M t r

I PROTOCOL CODES:

DOSE CODES: COL. 74

RESPONSE CODES: COL. 73

MAHWAV

M E R C K 00 00

7.

EGGERS E T A L .

Proprietary

Biological

and

Chemical

Data

89

ours, (3) documents already d i s t r i b u t e d t o the a p p r o p r i a t e members of the s c i e n t i f i c s t a f f , (4) magnetic tapes, o r (5) manual card f i l e s . We g e n e r a l l y accept b i o l o g i c a l data i n whatever form i s most convenient and ask that no one do any e x t r a work beyond making copies. Approximately 96 percent of the keypunching f o r entry t o the computer i s done w i t h i n the Biodata Department. Documents are u s u a l l y not s u p p l i e d w i t h card column headings. The data entry personnel are expected t o use a f a i r degree of judgment i n reading documents from which data are to be keyed. They are able t o convert from one dosage u n i t t o another, t o c a l c u l a t e percentages based on c o n t r o l v a l u e s , t o recognize v a r i o u s types of e r r o r s and omissions, and to use many other s k i l l s o b v i o u s l y w e l l above the l e v e l of the average keypunch operator. They a l s o type, take d i c t a t i o n , operate e l e c t r o n i c data p r o c e s s i n g equipment, use microform and sometimes w r i t e compute elementary nature. Some data a r r i v e at the department already on punched cards or a magnetic tape. I n t h i s case, the conversion t o the des c r i b e d format, p l u s the t r a n s f e r i n t o the master computer f i l e i s handled by members of the Biodata s t a f f . Each computer-stored record has l i n k s t o the input which produced i t . The date and t e s t number are the primary l i n k s , but, t o the extent p o s s i b l e , experiment number and/or notebook and page number are a l s o recorded. The t e s t number p l u s the p r o t o c o l code serve to reference the f i l e of p r o t o c o l s . An e l a b o r a t e a u d i t t r a i l i s kept f o r changes which need to be made i n records. A l l accessory f i l e s are created w i t h i n the department. Data are input i n t o computer storage on a d a i l y b a s i s i n batch mode, and a l l data are a c c e s s i b l e as soon as entered. This d a i l y input creates a monthly f i l e , which seldom runs over 30,000 records. At the end of each month, i t i s merged i n t o the total f i l e . The f i l e has been defined and the data records have been acquired. The next s e c t i o n deals w i t h how they are used. Search S t r a t e g i e s and Examples. We have the c a p a b i l i t y of (a) s e l e c t i n g a group of data records such as those keyed by the same compound number, t e s t number, date, combinations t h e r e o f , or other c o n d i t i o n s , (b) comparing those records according t o some defined c o n s t r a i n t s and l o g i c , and (c) using the r e s u l t i n g a n a l y s i s to r e t r i e v e f o r p r i n t i n g the a p p r o p r i a t e records from the same or other f i l e s . I n complex s i t u a t i o n s , p r e v i o u s l y w r i t t e n programs a s s i s t the process. The queries d i r e c t e d t o Biodata f a l l f a i r l y n e a t l y i n t o e i g h t c a t e g o r i e s (see F i g u r e 6). U s u a l l y some i n t e r f a c e w i t h one or more of the Chemdata f i l e s i s necessary. The connecting l i n k i s the i d e n t i f i c a t i o n number of the m a t e r i a l , the L-number.


90

RETRIEVAL

OF

MEDICINAL CHEMICAL

INFORMATION

I t i s important that the question be phrased so that i t i s completely unambiguous. We l i k e to negotiate each search i n d i v i d u a l l y w i t h the requesting s c i e n t i s t . I t i s important to the progress of the research that he gets what he needs, not n e c e s s a r i l y what he s t a r t e d out asking f o r . There are o p e r a t i o n a l programs which are o f t e n used f o r searches of types 2, 3, 4 and 6 (see Figure 6). No changes may be made i n output headings or formats and, except f o r type 6, no response, dose, or p r o t o c o l c o n d i t i o n s may be imposed. Names of compounds are always added, and we may ask that the computer i n c l u d e s t r u c t u r a l diagrams. I f a d d i t i o n a l s o p h i s t i c a t i o n i s r e q u i r e d , a custom program w i l l be w r i t t e n . The f i r s t four types are probably s e l f - e x p l a n a t o r y . The t h i r d type i n Figure 6 i s most h e a v i l y used. At the end of each monthly merge, a l l 2.5 m i l l i o n records are m i c r o f i l m e d using COM (Computer Output M i c r o f i l m ) a microfilm reader/printer i n a h a l f hour. Suppose the question i s "What do we know about a compound numbered L-475,878?" The answer i s i l l u s t r a t e d i n F i g u r e 7. This t y p i f i e s the answer to the very f i r s t t h i n g we wanted to be able to do - to put a l l the data on a given m a t e r i a l together i n one l i s t i n g . Search type 4 i n Figure 6 i s o r d i n a r i l y requested i n order to review the performance of a p a r t i c u l a r set of compounds i n c e r t a i n t e s t systems. I t a l s o provides a l i s t of those compounds which were not t e s t e d i n the defined s e t . This l i s t can then be sent to Chemdata w i t h a request f o r samples to be submitted f o r the t e s t ( s ) i n q u e s t i o n , i f the asking s c i e n t i s t d e s i r e s . Search type 6 i n Figure 6 i s i l l u s t r a t e d by the h y p o t h e t i c a l c o r r e l a t i o n shown i n Figure 8. A l l compounds which have been examined i n two groups of t e s t s are compared as to s p e c i f i e d responses i n each group. Here 3,600 compounds were t e s t e d both i n v i t r o and i n v i v o . The data show that 87 percent of the compounds a c t i v e i n the t e s t tube were a c t i v e i n mice and that l e s s than 1 percent of the compounds i n a c t i v e i n the t e s t tube were a c t i v e i n mice. Such c o r r e l a t i o n s are u s e f u l i n developing cost e f f e c t i v e t e s t i n g programs, as w e l l as e s t a b l i s h i n g unexpected r e l a t i o n s h i p s . Most r e l a t i o n s h i p s are not so obvious as t h i s , and may r e q u i r e s t a t i s t i c a l a n a l y s i s . S i n g l e record s e l e c t i o n s (Type 5 i n Figure 6) imply that there i s no need to consult any other record to determine whether you wish to save the one at hand. An i n t e r r e c o r d comparison r e q u i r e s , on the other hand, that a set of records i s to be r e t r i e v e d only i f two or more records, e i t h e r i n that set or another, meet some defined c r i t e r i a . " A d d i t i o n a l c o m p l i c a t i o n s " (Type 8 i n Figure 6) covers a m u l t i t u d e of s i n s . The program may be so horrendous that the computer run must be s p l i t up i n t o s e c t i o n s . Many s u b f i l e s may need to be b u i l t . I n t e r f a c e may be necessary with the output of a s t r u c t u r e search from Chemdata.


EGGERS E T A L .

Proprietary

Biological

and

Chemical

Data

I n f e c t i n g organism and s t r a i n and animal species Route of a d m i n i s t r a t i o n of drug and challenge Injection site ED , L D Therapeutic Index, MIC Percent of c o n t r o l Side e f f e c t s - t h r e e - d i g i t code Organ weights, body weight Doses o r concentrations i n v a r i o u s u n i t s , as mg/kg, mcg/ml 50

5 0

Figure 4.

Examples of data captured

Test number f i l e ( a c c e s s i b l e by computer terminal) Index t o Data Sheet f i l e ( a c c e s s i b l e by computer t e r m i n a l ) Search s t a t i s t i c s f i l e (on punched cards) Organism c r o s s - r e f e r e n c e f i l e (on punched cards) Cross reference f i l e s f o r : L-numbers MK-numbers NCI-numbers any other numbers necessary Figure 5.

1· 2. 3. 4. 5. 6. 7. 8.

Other files

No data, e.g., only p r o t o c o l s , counts, e t c . 20 A l l data f o r a p r e v i o u s l y defined s e t of t e s t s 55 A l l data f o r a p r e v i o u s l y d e f i n e d s e t of compounds 134 A l l data f o r a p r e v i o u s l y d e f i n e d s e t of compounds i n a p r e v i o u s l y defined s e t of t e s t s 121 S i n g l e record s e l e c t i o n s , w i t h or without f i l e matching 70 Correlations 7 I n t e r r e c o r d comparisons 185 A d d i t i o n a l complications 22 Total 614 Number r e q u i r i n g s t r u c t u r e s : 148 Figure 6.

Types of searches released—1977


RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION


7.

EGGERS E T A L .

Proprietary

Biological

and

Chemical

Data

93

Except f o r o p e r a t i o n a l programs such as the four noted above, a l l questions i n v o l v e custom programs or manual searches or p o s s i b l y a combination of s e v e r a l methods. Requests to be repeated f a i r l y o f t e n w i l l be catalogued i n our search l i b r a r y . Roughly 65-70 percent of searches i n v o l v e one-time programs. The needs of researchers vary g r e a t l y , depending upon s u b t l e d i f f e r e n c e s which cannot be defined adequately u n t i l they begin to i n t e r a c t w i t h the data bases. Response t o t h i s c o n s t a n t l y changing need i s of utmost importance i n a research environment. The custom programming i s handled by members of the Biodata Department. There are four t e c h n i c a l people and f i v e c l e r i c a l people. A l l are p r e s e n t l y w r i t i n g computer search programs upon demand. These programs are assigned according t o t h e i r current degree of e x p e r t i s e . The 614 searches ru down i n t o the method use The t o t a l , 805, i s g r e a t e r than 614, because some r e q u i r e d more than one method. The 111 persons requesting these 614 searches were 46 chemists, 50 b i o l o g i s t s , and 15 from l e g a l , management, marketing and other i n f o r m a t i o n science areas. The term "The Data A n a l y z e r " i n Figure 9 r e f e r s t o a very e f f i c i e n t FORTRAN-based software package marketed by Program Products, Inc., Nanuet, New York. I t i s used f o r n e a r l y a l l the custom programs. Operations are mostly i n batch mode, although there i s a t e l e t y p e t e r m i n a l , which i s s u i t a b l e f o r subsets of the master f i l e and f o r some s m a l l e r , accessory f i l e s . Each of the 80 characters i n the record i s searchable, as i s any combination thereof. F i e l d s may be defined as r e q u i r e d by each search. S u b f i l e s of records o r p o r t i o n s of records may be created. Computations may be performed on a l l numeric f i e l d s , bearing i n mind that a numeric f i e l d i n one t e s t number may be alpha i n another. Comparing, counting, s o r t i n g are performed as necessary. The records i n computer storage may look very d i f f e r e n t from those presented t o the requesting s c i e n t i s t , depending upon the need of that s c i e n t i s t . The f i n a l output i s t a i l o r e d t o the needs of the s c i e n t i s t . The s i x records i n Figure 10 were combined w i t h names and s t r u c t u r e s from the Chemical Data F i l e s , reformatted, other i n f o r m a t i o n added, and a report produced as shown i n F i g u r e 11. This i s an example of what the s c i e n t i f i c s t a f f i s p r e s e n t l y r e c e i v i n g . The query here was: Produce a report of l a s t week's r a t hypertension data using a p p r o p r i a t e headings f o r each compound i n t h i s order ( l e f t t o r i g h t ) any combining compound number, r a t type, onset time, dose, route of a d m i n i s t r a t i o n , number of r a t s , codes f o r blood pressure change, d u r a t i o n of blood pressure change, heart r a t e change, and date. Omit book and page r e f e r e n c e , t e s t number and v e h i c l e code and add an expanded t e s t t i t l e as a heading. Define a l l codes on each page of


RETRIEVAL

A c t i v e i n Mice and A c t i v e i n Test Tube 153 Compounds

O F MEDICINAL

CHEMICAL

INFORMATION

| | |

A c t i v e i n Mice and I n a c t i v e i n Test Tube 19 Compounds

I n a c t i v e i n Mice and | A c t i v e i n Test Tube | 27 Compounds |

I n a c t i v e i n Mice and I n a c t i v e i n Test Tube 3401 Compounds

Of the 180 (153 + 27) compounds a c t i v e i n t e s t tube, 153 (or 87%) are a l s o a c t i v e i n mice. Of the 3420 (340 t e s t tube, 340 mice. Figure 8.

Figure 9.

Correlation study (hypothetical)

Method of solution

Hand Look Up Personal Contact EDP Machine Terminal O p e r a t i o n a l Program Microform The Data Analyzer S t r u c t u r e Driven Other Total

103 4 46 3 172 179 290 3 5 805

62734600L012725H Y PERTEN. RAT

72-15(2) 112

20.

MG/KG5332 121672

62734600L012725HYPERTEN. RAT

7 2 - 1 5 U 112

20.

MG/KG

62734600L012725H Y PE RTEN. RAT

73- 39( ) 112

. 08MG/KG0004 011373 . 02MG/KG0004 011373

92 121672


73- 38(2) 112

62734600L012725H Y PE RTEN. RAT

73- 38(

) 112 06.0

. 02MG/KG0214 011373


73-36(

) 112 01.0

. 08MG/KG0314 011373

Figure 10.

Records in computer storage


7.

EGGERS E T A L .

Proprietary

Biological

and

Chemical

Data

95

the r e p o r t . Add names and s t r u c t u r e s . D i s t r i b u t e 19 copies to the l i s t of persons designated by Dr. Sweet, who w i l l w r i t e a one page summary cover memo. (This p a r t i c u l a r report was 90 pages long.) Another type of question i n v o l v e s no data d i s p l a y a t a l l , but can i n v o l v e j u s t as much programming. An example i s shown i n F i g u r e 12. The requestor now can r e a d i l y see t h a t , of the 223 compounds t e s t e d , 9 produced a response code of 3, and 1 produced a response code of 1. Note that t h i s may be only 9 compounds, s i n c e o b v i o u s l y there was more than 1 t e s t per compound. Another example might i n v o l v e a new t e s t i n c h i c k s and there are no g u i d e l i n e s except, perhaps, a t i g h t budget. What compounds are there f o r which a p h y s i c a l sample e x i s t s (at l e a s t 250 mg), which were not t o x i c t o c h i c k s at 0.1% i n the d i e t ? The answer t o t h i s questio c h i c k s t r y i n g t o f i n d non-toxi Chemical Data The Chemdata Systems at Merck Sharp & Dohme Research L a b o r a t o r i e s have grown over the past dozen years from a few manual f i l e s to a complete network of i n t e r f a c e a b l e , searchable data bases. For the purpose of comparison, the o r i g i n a l systems c o n s i s t e d of: (1) a chemical s t r u c t u r e f i l e which u t i l i z e d the c l a s s i f i c a t i o n system which was developed by Dr. F. W i s e l o g l e (1) i n 1946; (2) a manual index card f i l e of the sample r e p o s i t o r y ; (3) t r a n s a c t i o n summary sheets c o n t a i n i n g manually posted t r a n s m i t t a l s and; (4) a chemical name f i l e which was a p o r t i o n of the record of the t r a n s a c t i o n summary sheets. Our current computerized i n d i v i d u a l systems, i n c l u d i n g the o r i g i n a l design, content, search and i n t e r f a c e c a p a b i l i t i e s w i t h other systems, are d e s c r i b e d , f o l l o w e d by the enhancements that were added as we gained experience from user i n t e r a c t i o n . An overview summarizes e x a c t l y how the research s c i e n t i s t at MSDRL u t i l i z e s the systems and how he has r e a l i z e d b e n e f i t s therefrom. Chemical Structure Information Systems (CSIS). The most important system i s our Chemical S t r u c t u r e Information System. The d e t a i l s of t h i s system have been published ( 2 ) . However, some major h i g h l i g h t s deserve review. In t h i s system both the updating and the querying of the chemical s t r u c t u r e f i l e begin w i t h t y p i n g the s t r u c t u r e s on a Magnetic T a p e / S e l e c t r i c Typewriter equipped w i t h a t y p i n g e l e ment bearing appropriate bonding c h a r a c t e r s . The tapes a r e processed by computer i n batch mode. For f i l e updates the s t r u c t u r e s are subjected t o numerous v a l i d i t y checks and are analyzed t o create a connection t a b l e and a s s i g n b i t screen elements. Before entry i n t o the master f i l e , newness t o the f i l e i s e s t a b l i s h e d . The e n t i r e i n p u t


RETRIEVAL

96

Antihypertensive

Evaluation Codes*

O F MEDICINAL

1 · 20-29 mm Deer

Report No. 256

2 · 30-39 mm Deer

INFORMATION

Heart Rate

Blood Press.

Blood Press. Eff.

Rat Assay

CHEMICAL

1 · Ouest. Chanqe

Duration 1 - Lt 4 hrs

2 - 60-100 Β / M i n e r

3 · GT 40 mm Deer

2 - 4-7

hrs

3 - CT100 B / M I n c r

8 * G T 2 0 m m Deer

3 -GT7hrs

5 - GT 100 B/M Deer

9 · Toxic 0 · No Effect

L-627,346-00L Mol. Wt.

C17H22F3N3O2

357.37

NS Baldwin,

J.J.

M1536-1105-1V

2-[4-(3-tertbutylamino-2-hydroxypr(^

Onset

0 •P.O.

Rat

Time

Dose

Β Τ

Type

Hrs.

Mg/Kg

01

SH

01

SH

01

SH

01

SH

01

SH

CI

SH

Blood

Blood Press. Duration

1 •I. P. 2 •I.V.

Ν α of

20.

1

2

20.

1

9

.08

1

0

0

.02

1

0

06.0

.02

1

01.0

.08

1

Rats

2

Figure 11.

Press. Code

Heart Rate

Code

Code

3

5

Mo Da 12

16

12

16

0

01

13

0

0

01

13

1

2

0

01

13

1

3

0

01

13

3

Final report

408 = Count of t e s t s f o r the p e r i o d 223 = Count of compounds 1 1 0 0 15 9

= = = = = =

Count Count Count Count Count Count

of t e s t s f o r response code 1 of compounds f o r response code 1 of t e s t s f o r response code 2 of compounds f o r response code 2 of t e s t s f o r response code 3 of compounds f o r response code 3

Figure 12.

Bimonthly report for Test X


7.

EGGERS E T A L .

Proprietary

Biological

and

Chemical

97

Data

r e c o r d , which i n c l u d e s other d e s c r i p t i v e i n f o r m a t i o n , i s s t o r e d i n the computer. Figure 13 represents a chemical s t r u c t u r e record as i t i s input t o the system. The L-number i n the upper l e f t i s the r e g i s t r a t i o n number. This i s f o l l o w e d by the molecular formula, then any stereo d e s c r i p t o r s , reference t o source, and f i n a l l y the chemical s t r u c t u r e . This i s a t y p i c a l s i n g l e record. The f i l e contains approximately 135,000 such records. The year of r e g i s t r a t i o n i s added t o the record a u t o m a t i c a l l y at the time of i t s input t o the system. Tapes bearing s u b s t r u c t u r e queries are processed s i m i l a r l y through e d i t and automatic b i t screen assignment. Although answers may be s e l e c t e d as the r e s u l t of b i t matching alone, t h e search procedure i s normally allowed t o proceed past t h i s stage to the atom-by-atom comparison. Figure 14 l i s t s th capabilities. Figure 15 l i s t s the search parameters which may be invoked i n order t o provide greater s p e c i f i c i t y f o r search q u e r i e s . The CSIS system, as i l l u s t r a t e d i n Figure 16 e x t r a c t s a n c i l l a r y data from the chemical name f i l e , the sample r e p o s i t o r y f i l e and the b i o d a t a f i l e . I n the case of b i o d a t a , i t may be s p e c i f i e d that "ALL" b i o d a t a be p r i n t e d f o r the s p e c i f i c compounds r e t r i e v e d . A d d i t i o n a l c r i t e r i a such as dose and/or a c t i v i t y l e v e l requirements may be i n c l u d e d . The system i s capable of i n t e r r o g a t i n g a customized d i c t i o n a r y of b i o l o g i c a l t e s t t i t l e s s i n g u l a r l y or i n combination, which when n u m e r i c a l l y designated, p r i n t s only those t e s t r e s u l t s stored w i t h i n that d e f i n i t i o n . This data base i s searched n i g h t l y i n batch mode and i s updated on a weekly b a s i s . The s t r u c t u r e f i l e i s capable of being l i n k e d w i t h other MSDRL f i l e s by p r o v i d i n g s t r u c t u r a l i d e n t i f i c a t i o n f o r r e p o r t s upon request throughout MSDRL. Chemical Names. Figure 17 i l l u s t r a t e s the format of the Chemical Name F i l e . Each record contains the r e g i s t r y L-number f o l l o w e d by the chemical name which i n most cases observes CA g u i d e l i n e s . The length of the name i s l i m i t e d t o 180 c h a r a c t e r s . This f i l e i s searchable by L-number and i n some cases by t e x t r e c o g n i t i o n of p o r t i o n s of the name u t i l i z i n g programs w r i t t e n by the departmental s t a f f . Compound Repository. Figure 18 represents the record format of a compound r e p o s i t o r y r e c o r d . The r e g i s t r y number i s f o l l o w e d by the l o c a t i o n d e s c r i p t o r s of each s p e c i f i c sample. The f i l e i s searchable by L-number and contains 125,000 records sequenced i n L-number order. Transaction F i l e .

The most r e c e n t l y developed

system i s


RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

L-590,226-00A C H C1N0 NS WITZEL, Β. M5247-118-5 19

Figure 13.

16

4

CSIS input record

Cl

Total Structure Fragment S t r u c t u r e Multifragment S t r u c t u r e L-number ( S p e c i f i c or Range) Year Plus Fragment S t r u c t u r e Reference Plus Fragment S t r u c t u r e Exact Molecular Formula Plus Fragment S t r u c t u r e Molecular Formulae (Minimum or Exact) Figure 14.

CSIS retrieval capabilities

Statements of i n d e f i n i t e s u b s t i t u e n t s (X's) a t i n d e f i n i t e p o s i t i o n s (Z's) Ring (R) and A c y c l i c (A) d e c l a r a t i o n s I n d e f i n i t e bonds - dotted Apostrophe - r e s t r i c t s s u b s t i t u t i o n Carbon dot - p r o h i b i t s s u b s t i t u t i o n Valence d e c l a r a t i o n s Charge d e c l a r a t i o n s Boolean Logic (AND, OR ( i n c l u s i v e ) , BUT NOT) Repeating u n i t s Abbreviations (Any, None, H a l , Peptides) L i m i t e d r e t r i e v a l volume 500, substructure search Figure 15.

CSIS search variables


7.

EGGERS

Proprietary

ET AL.

Biological

and

Chemical

Data

L-590,226-OOA C H C 1 N 0 Mol. Wt. 357.796 NS WITZEL, Β. M5247-118-5 1 9

1 6

4

CH COOH 2

INDOMETHACIN

L-NUMBER

SALT

590,226 590,226

00 00

2135 2135

A A BATCH 39 47

DESCRIPTION ANALGESIC IN RATS 2175 ANALGESIC IN RATS 2176

LOCATION

CONTAINER

TRAY-0242 TRAY-0249

BOTTLE-036 BOTTLE-017

Figure 16.

L-590,226 OOA

TEST NUMBER

BATCH

CSIS computer report

INDOMETHACIN

L-590,226 04J N,N*-DIBENZYLETHYLENEDIAMINE SALT OF l - ( 4 CHLOROBENZOYD-2-METHYL-5-METHYL-5-METHOXYINDOLE-3-ACETIC ACID L-590,226 07R POTASSIUM l-(4-CHLOROBENZOYL)-5-METHOXY-2METHYL-3-INDOLEACETATE L-590,226 09V MONOBASIC ALUMINUM l-(4-CHLOROBENZOYL)-5METHOXY-2-METHYL-3-INDOLEACETATE L-590,226 16T MIX INDOMETHACIN WITH LACTOSE Figure 17.

Chemical name file


99

100

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

the Transaction F i l e Data Base. Figure 19 shows the format of a t y p i c a l record. This system has replaced v i r t u a l l y a l l of the manual record maintenance i n the Chemical Data Department. I t i s r a p i d l y becoming one of the most important systems. The record contains the L-number, date, p r o j e c t and t e s t , sample s i z e , source and requested t e s t i n g a c t i o n to be performed. A l l of these f a c e t s are searchable. This system i s r o u t i n e l y i n t e r f a c e d w i t h the Chemical S t r u c t u r e Information System i n order to provide compound p r o f i l e data f o r users. This f i l e not only gives an h i s t o r i c a l record f o r each compound, but a l s o provides computer generated d e l i v e r y forms w i t h complete i d e n t i f i c a t i o n , which accompany the samples to the t e s t i n g s i t e . This f i l e i s updated d a i l y . Chemdata Enhancements s t r u c t u r e search i s requested most notably b i o l o g i c a l . The o r i g i n a l CSIS design r e q u i r e d the p r i n t i n g of every s t r u c t u r e r e t r i e v e d from a search r e g a r d l e s s of other search c r i t e r i a . In order to accommodate a l l search c r i t e r i a and yet optimize the u t i l i t y of search r e p o r t s , s e v e r a l programs were developed which b a s i c a l l y provide a p r i n t o p t i o n . This o p t i o n i s invoked at the completion of CSIS searches. In e f f e c t , the o p t i o n i s used to r e s t r i c t the chemical answers p r i n t e d from a CSIS search to those w i t h or without b i o l o g i c a l data and/or w i t h or without sample r e p o s i t o r y data. Figure 20 i l l u s t r a t e s the commands which have been incorporated as part of the CSIS search question. Transaction F i l e / I n v e n t o r y . The next major enhancement, c u r r e n t l y under development, i s the Chemical Sample Inventory System which i s part of the o p e r a t i o n a l Transaction F i l e Data Base. This system provides the researcher w i t h an exact amount of compound a v a i l a b l e from the Sample Repository. Knowledge of t h i s i s e s s e n t i a l to those r e s p o n s i b l e f o r overseeing and s e l e c t i n g of compounds f o r both chemical and b i o l o g i c a l t e s t i n g . Without t h i s c a p a b i l i t y , the s e l e c t i o n process becomes d i f f i c u l t ; s u f f i c i e n t s u p p l i e s may or may not e x i s t . I f adequate s u p p l i e s of a chosen candidate do not e x i s t , the process must be repeated and the net e f f e c t i s a c o s t l y waste of research time and t e c h n i c a l manpower. F i l e b u i l d i n g was s t a r t e d by net weighing every new compound submitted to the Sample Repository f o r c a t a l o g u i n g and storage. Concurrently, samples already i n storage were weighed and the net weight data added to the Transaction F i l e . Figure 21 shows how the record looks w i t h t h i s a d d i t i o n a l data. The inventory data are recorded as of a s p e c i f i c date. One of our s t a f f members has w r i t t e n a program which now provides current net weight v a l u e s . This i s accomplished by s u b t r a c t i n g


7.

EGGERS E T A L .

Proprietary

Biological

and

Chemical

Data

101

COMPOUND LIST AND LOCATION 01/05/78 W. GALL FOR ANTIINFLAMMATORY STUDIES

SEARCH NO. 010377 L-NUMBER

LOCATION

CONTAINER

TRAY-0752 TRAY-0752 TRAY-0752 TRAY-0752 TRAY-0752 SHLF-0011 SHLF-0011

BOTTLE-049 BOTTLE-014 BOTTLE-010 BOTTLE-048 BOTTLE-008 BOTTLE-022 BOTTLE-015

590,226-OOA

Figure 18.

INDOMETHACIN L-590,226-OOA C

1 9

H

1 6

C1N0

4

M o l . Wt.

357.796

NS WITZEL,

Β.

M5247-118-5

CI

BATCH

PROJECT

NO.

NO.

00

00

0001

0001

TEST

DATE

NO.

MO D A Y R

0000

01 18 72

0000

0 1 18 72

SAMPLE SIZE 0000.1000

0005.0000

UNITS GRAM

GRAM

SOURCE DC

DC

COMMENTS F A I N BROWN

ISLAND

BATCH

104

SANDBERG, BATCH

00

0001

0000

0 1 18 72

0010.0000

GRAM

DC

SWEDEN

104

ARMSTRONG, BATCH

Figure 19.

UNIVERSITY,

RHODE

104

Transaction file record


CANADA

RETRIEVAL

102

O F MEDICINAL

CHEMICAL

INFORMATION

Β - P r i n t only compounds which do have a l l the requested types of b i o d a t a . NB - P r i n t only compounds which have none of the requested types of b i o d a t a . S - P r i n t only compounds which have sample collection location. NS - P r i n t only compounds which do not have sample l o c a t i o n s . I f both b i o d a t a and sample codes are present, a connective "AND" i s assumed. Figure

INDOMETHACIN L-590.226-OOA C

1 9

H

1 6

C1N0

4

M o l . Wt.

357.796

NS WITZEL,

BATCH NO. 01

PROJECT

TEST

DATE

NO.

NO.

MO D A Y R

0005

1 0 20 77

0000

Β.

M5247-118-5

SAMPLE SIZE 0005.1650

UNITS GRAM

SOURCE SC

COMMENTS INVENTORY/DATE ADDITION

01

1064

0000

1 0 26 77

0000.0050

GRAM

SC

DOUGHERTY FOR FE-SOD

09

2320

9325

1 0 26 77

0000.0200

GRAM

SC

ASSAY

DULANEY FOR IN VITRO

Figure 21.

1482-28

BY PATCHETT

CHEMOTHERAPY

Transaction file/inventory report


7.

EGGERS E T A L .

Proprietary

Biological

and

Chemical

Data

103

the weights of samples that have been dispensed from a s p e c i f i c sample a f t e r the inventory weight was recorded. This weight i s then a u t o m a t i c a l l y l i s t e d as the "Net Inventory". See F i g u r e 22. In summary, the "Net Inventory" System coupled w i t h CSIS search r e t r i e v a l , the t r a n s a c t i o n p r o f i l e and b i o d a t a r e s u l t s provides the researcher w i t h the d e s i r e d overview and helps t o p l a n f u t u r e research a c t i v i t i e s . In reviewing the Chemdata Searches completed during 1977 s t a t i s t i c s were compiled as shown i n F i g u r e 23. I t i s obvious f o r one t o conclude that the b i g use of these systems i n v o l v e s the c a t e g o r i e s : (1) s u b s t r u c t u r e search w i t h sample a v a i l a b i l i t y data, and (2) s u b s t r u c t u r e search w i t h b i o l o g i c a l and sample a v a i l a b i l i t y data. Interaction The researcher i s using the systems to uncover new pathways of research by a n a l y z i n g and c o r r e l a t i n g data that are s u p p l i e d from the aforementioned f i l e s . Given an " i d e a " about a s t r u c t u r e / a c t i v i t y r e l a t i o n s h i p , questions a r i s e : 1. Has compound A been i n Test B? 2. I f so, what were the r e s u l t s ? 3. I f not, do we have a sample i n the c o l l e c t i o n ? And 4. i f so, i s there enough m a t e r i a l t o conduct the t e s t ? These systems are becoming more and more important as they grow and i n c l u d e newer research e f f o r t s . Figure 24 i l l u s t r a t e s the i n t e r a c t i o n among the v a r i o u s f i l e s of both b i o l o g i c a l and chemical data which i s now f a i r l y r o u t i n e at MSDRL. The f i r s t two c o n d i t i o n s r e q u i r e examination of the b i o l o g i c a l data. A p o s i t i v e match w i t h the sample a v a i l a b i l i t y f i l e plus a negative match w i t h the s t r u c t u r e c o n s t r a i n t s produce a l i s t of candidate compounds. This l i s t then can be fed back i n t o the Biodata F i l e , the Chemname F i l e , and the S t r u c t u r e F i l e to produce the d e s i r e d r e p o r t . Suppose the answer encompasses 150 compounds and 6 of them look promising on t h i s b a s i s . A s y n t h e t i c program to make analogs may be launched. A new product may r e s u l t .


RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

INDOMETHACIN L-590,226 00A C H C1N0 Mol. Wt. 357.796 NS 1 9

1 6

4

W i t z e l , Β. M5247-118-5 01

Net Inventory

5.1400 Gram

Figure 22.

Tray-1208

Bottle-004

Net inventory

CSIS Search T o t a l 1. 2.

T o t a l number of searches conducted T o t a l number of i n d i v i d u a l users A. 74 chemists B. 3 patent attorneys C. 5 biologists D.

695 86

4 other i n f o r m a t i o n departments

Types of Searches 1.

Substructure A. s u b s t r u c t uTypes r e + sample a v a i l a b i l i t y data B. s u b s t r u c t u r e + b i o d a t a C. s u b s t r u c t u r e + b i o d a t a + sample availability D. s u b s t r u c t u r e - compound i d e n t i f i c a t i o n only Registration Identification A. L-number r e g i s t r a t i o n + sample availability B. L-number r e g i s t r a t i o n s + b i o d a t a C. L-number r e g i s t r a t i o n s + b i o d a t a + sample a v a i l a b i l i t y D. L-number r e g i s t r a t i o n s ( i d e n t i f i c a t i o n only) P h y s i c a l Inventory Searches (Operational Nov. 1977) A. L-number + b i o d a t a + p h y s i c a l inventory B. L-number + p h y s i c a l inventory (2 searches Figure 23.

MSDRL—use

i n v o l v i n g 150 compounds) of the Chemdata systems during 1977


216 28 313 14

19 28 49 28

100

EGGERS

E T AL.

Proprietary

Biological

and Chemical

Data

Are there any compounds which have: An e l e c t r o s h o c k p r o t e c t i o n ED^Q of 2 0 0 . 0 mg/kg, both i n mice, and f o r which we have an a v a i l a b l e sample of a t l e a s t 1 5 0 mg. Given the above c o n d i t i o n s , please d i s p l a y a l l i n v i v o data i n computer storage, together w i t h chemical names and s t r u c t u r e s , but e x c l u d i n g compounds having 7-membered r i n g s , w i t h 5 carbons and 2 n i t r o g e n s 1,4 t o each other. Figure 24.

Biodata/Chemdata interaction to solve query


106


Acknowledgments Our thanks go to Mr. T. Boyer, A s s i s t a n t D i r e c t o r of Merck Sharp & Dohme Research L a b o r a t o r i e s Information Systems & Programming, Mr. R. T. F o r d , Manager, Research Data P r o c e s s i n g , Mr. W. L . H e n c k l e r , Senior Chemical Information S p e c i a l i s t of the Chemical Data Department, Mr. W. C. Kinahan, S u p e r v i s o r , S c i e n t i f i c Information A n a l y s t of the B i o l o g i c a l Data Department, Mr. F . W. Landgraf, now of Mobay Chemical C o . , Mr. C. J . M i l l e r , Systems P r o j e c t Supervisor and Mr. W. P a t e r , D i r e c t o r of Computer Operations. We a l s o wish to thank many s t a f f members of Merck & C o . , I n c . f o r t h e i r part i n system development. P a r t i c u l a r thanks go to Nancy N i k i p e r and Ann D e N i t t i s f o r a l l the t y p i n g i n v o l v e d . Abstract An open-ended computer system i s described f o r the c o l l e c t i o n , s t o r a g e , r e t r i e v a l and dissemination of b i o l o g i c a l d a t a . Facets of the system i n c l u d e v a r i a b l e u n i t record formats w i t h defined f i e l d s derived from s i g n i f i c a n t aspects of t e s t protocols. I n t e r f a c e i s commonly made w i t h other data bases mentioned below, s i n g l y or i n combination, i n c l u d i n g chemical structure constraints. The common l i n k to the other data bases i s the compound r e g i s t r a t i o n number. Output d i s p l a y may i n c l u d e records from any or a l l of the data bases accessed. Counts, r e s u l t s of computation, or t a b l e s may be i n c l u d e d , as requested by members of the s c i e n t i f i c s t a f f . The Chemical S t r u c t u r e Information System has been p a r t i a l l y described i n 1976 ( 2 ) . The complete network of the chemical and b i o l o g i c a l information systems, i n c l u d i n g Sample Repository Data Base, Transaction F i l e Data Base, as w e l l as Chemical Name Data Base, i s d e s c r i b e d . Searches performed during a recent calendar year on Biodata and Chemdata Systems have been analyzed, and the r e s u l t s are discussed w i t h emphasis on the i n t e r p l a y between the chemical s t r u c t u r e , s u b s t r u c t u r e and b i o l o g i c a l data segments of the o v e r a l l system. Literature

References

1.

Emmett L . Buhle, E l i n o r D. Hartnell, Alexander M. Moore, Louise R. W i s e l o g l e , and F . Y . W i s e l o g l e "A New System f o r the Classification of Compounds: A C o n t r i b u t i o n from the Survey of A n t i m a l a r i a l Drugs", J. Chem. Ed., 23., 375 (1946).

2.

Horace D. Brown, Marianne Costlow, Frank A . C u t l e r , Jr., A l b e r t N . Demott, Walter B . Gall, David P. Jacobus, and Charles J. Miller "The Computer-Based Chemical Structure Information System of Merck Sharp & Dohme Research L a b o r a t o r i e s " , J . Chem. I n f . and Computer Sci., 16, 5 (1976).



8

Progress toward an On-Line Chemical and Biological I n f o r m a t i o n S y s t e m at t h e

Upjohn

Company

W. J. HOWE and T. R. HAGADONE The Upjohn Company, Kalamazoo, MI 49001

Over the past t e n a c e u t i c a l i n d u s t r y have way computers are utilized in research and research support functions. E a r l y a p p l i c a t i o n s tended to focus on numerical tasks such as statistical analyses and quantum mechanical c a l c u l a t i o n s o r on the a r c h i v a l storage of i n f o r m a t i o n r e l a t e d to the chemistry o r b i o l o g y of research substances. I n the latter case, i n f o r m a t i o n retrieval systems were often u n w i e l d l y and r e q u i r e d c o n s i d e r a b l e e x p e r t i s e f o r their use. The l a b o r a t o r y researcher u s u a l l y had to work through an intermediary in order to r e t r i e v e i n f o r m a t i o n from such systems. More r e c e n t l y , we have seen a shift of emphasis to where computers are now recognized as i n d i s p e n s a b l e t o o l s in the day-to-day o p e r a t i o n of scientific research. On-line i n t e r a c t i v e methods have placed the i n f o r m a t i o n resource much c l o s e r to the end u s e r . I n a d d i t i o n to their "traditional" a p p l i c a t i o n s , computer-based systems are being employed to assist in the design of organic syntheses, in the interpretation of s p e c t r o s c o p i c d a t a , i n the design and development of new drug candidates, f o r r e a l - t i m e experiment c o n t r o l , and in a wide variety of r e l a t e d areas (1-6)· The retrieval and m a n i p u l a t i o n of m e d i c i n a l chemical i n f o r m a t i o n is another area in which computer—based systems have made an i m pact and which will become i n c r e a s i n g l y important in future y e a r s . This paper w i l l focus on a p r o j e c t which has been under way at the Upjohn Company to develop a comprehensive chemical and b i o l o g i c a l i n f o r m a t i o n system to be used by research s c i e n t i s t s and research support p e r s o n n e l . C a p a b i l i t i e s of the system w i l l event u a l l y i n c l u d e o n - l i n e s t r u c t u r e r e g i s t r y , s t r u c t u r e and s u b s t r u c ture s e a r c h i n g , the r e t r i e v a l and m a n i p u l a t i o n of pharmacological t e s t data, and the r e t r i e v a l of s p e c t r o s c o p i c , p a t e n t , and other types of s t r u c t u r e - a s s o c i a t e d d a t a . There are c u r r e n t l y a number of systems i n the company which are being used f o r the storage of b i o l o g i c a l data a s s o c i a t e d w i t h compounds that have been s y n t h e s i z e d f o r s c r e e n i n g . I n most cases, the o p e r a t i o n of these systems has i n the past been q u i t e i n d e pendent of " c h e m i c a l l y - o r i e n t e d " i n f o r m a t i o n . Chemical s t r u c t u r e 0-8412-0465-9/78/47-084-107$06.25 Published 1978 American Chemical Society In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

RETRIEVAL

108

OF

MEDICINAL

CHEMICAL

INFORMATION

and s u b s t r u c t u r e searching has been accomplished through the use of a fragment code which was developed i n the l a t e 1950's, and which, d e s p i t e a number of drawbacks that are commonly inherent i n fragment-based systems, has met the needs of our s c i e n t i s t s f o r a number of y e a r s . However, s i n c e one of the goals of the new i n formation system i s to p r o v i d e a means f o r i n t e r a c t i v e l y a c c e s s i n g both the chemical s t r u c t u r e data and a s s o c i a t e d pharmacological data, and f o r the e x t r a c t i o n of subgroups of compounds which could then, f o r example, a c t as source data f o r end-user a p p l i c a t i o n s such as p a t t e r n r e c o g n i t i o n a n a l y s e s , the design of a f l e x i b l e and e f f i c i e n t chemical s t r u c t u r e entry and search system became the i n i t i a l t a r g e t of our a t t e n t i o n . The chemical s t r u c t u r e system c o n s i s t s of three p a r t s , i n d i f f e r e n t stages of development: (a) the s t r u c t u r e database, a c o l l e c t i o n of approximately 60,000 chemical s t r u c t u r e t i o n of which has (b) the s t r u c t u r e entry system, an i n t e r a c t i v e computer-graphics based system which was developed t o create the i n i t i a l d a t a base; p o r t i o n s of t h i s w i l l a l s o be i n c o r p o r a t e d i n the compound r e g i s t r y and search system, (c) the compound r e g i s t r y and search system, c u r r e n t l y under development, which c o n s i s t s of two p a r t s : (1) an o n - l i n e r e g i s t r y f a c i l i t y which w i l l a l l o w i n t e r a c t i v e d a i l y updating of the database, and, (2) the query f a c i l i t y , which w i l l a l l o w o n - l i n e i n t e r a c t i v e s t r u c t u r e and s u b s t r u c t u r e searching and e v e n t u a l searching and m a n i p u l a t i o n of a s s o c i a t e d pharmacological i n f o r m a t i o n . The system w i l l enable the user t o d i s p l a y the r e t r i e v e d i n f o r m a t i o n i n a convenient format and to produce h i g h q u a l i t y hard copy output of both s t r u c t u r a l and t e x t u a l data. 1.

The S t r u c t u r e Database

A key phase of the p r o j e c t i n v o l v e d the c r e a t i o n of the s t r u c t u r e database, a g r a d u a l l y e n l a r g i n g c o l l e c t i o n of a p p r o x i mately 60,000 chemical s t r u c t u r e s which over the years had e i t h e r been synthesized in-house f o r t e s t i n g purposes o r obtained from o u t s i d e o r g a n i z a t i o n s . The fragment-coded search system a l s o operated on t h i s c o l l e c t i o n of compounds; however, s i n c e fragment codes represent s t r u c t u r a l a t t r i b u t e s , the codes could not be used to regenerate complete connection t a b l e s . A s t r u c t u r e entry system was designed which, by using computer g r a p h i c s as the input medium, would a l l o w d i r e c t t r a n s c r i p t i o n of the s t r u c t u r e diagrams from hard copy format i n t o the computer system. Connection t a b l e s would be generated i n r e a l - t i m e as the s t r u c t u r e drawing o p e r a t i o n progressed. The s t r u c t u r e e n t r y program was ready f o r use about 1-1/2 years ago and f u l l s c a l e s t r u c t u r e e n t r y began a t the s t a r t of 1977. Although many


8.

HOWE AND HAGADONE

On-Line

Information

System

109

e r r o r - d e t e c t i o n devices were b u i l t i n t o the system, there were s t i l l c e r t a i n types of e r r o r s which could s l i p by and enter the data base. For that reason i t was decided at the outset t h a t each s t r u c t u r e would have to be entered t w i c e , by d i f f e r e n t t e r m i n a l operators, thereby enabling an i d e n t i t y check to be performed on the host computer. E r r o r checking by manual comparison of each entered s t r u c t u r e w i t h a hard copy record would, i t was f e l t , take j u s t as long as i t would take to redraw a s t r u c t u r e a second time and would s t i l l provide no guarantee that a l l e r r o r s had been caught. The s t r u c t u r e entry o p e r a t i o n has j u s t r e c e n t l y been comp l e t e d , having taken c o n s i d e r a b l y l e s s time than o r i g i n a l l y a n t i c i p a t e d . Now that the high volume input of the database "backlog" i s done, i t i s planned that r o u t i n e d a i l y update of the database w i t h low volume " c u r r e n t " s t r u c t u r e s w i l l be handled by the onl i n e r e g i s t r y f a c i l i t y whic be discussed l a t e r . 2.

The S t r u c t u r e Entry System

The s t r u c t u r e entry system was designed to accommodate r a p i d e r r o r - f r e e s t r u c t u r e e n t r y , w i t h much c o n s i d e r a t i o n given t o s t r u c t u r e diagram cosmetics. I t was a l s o designed so that i t could be e a s i l y i n c o r p o r a t e d i n t o the compound r e g i s t r y and search system w i t h l i t t l e or no m o d i f i c a t i o n . For that reason, we w i l l present an o p e r a t i o n a l overview of the g r a p h i c a l s t r u c t u r e e n t r y system, f o c u s i n g i n p a r t i c u l a r on i t s use i n the c r e a t i o n of the s t r u c t u r a l database. (a) Hardware. The data entry t e r m i n a l i s operated essent i a l l y as a stand-alone computer system (Figure 1) which t r a n s mits completed s t r u c t u r e connection t a b l e s to the host machine (370/155) where they are compared a g a i n s t t h e i r d u p l i c a t e s t r u c tures (double e n t r y ) . Once a day an e r r o r l o g i s p r i n t e d to enable c o r r e c t i o n of s t r u c t u r a l e r r o r s (using a s i m i l a r program on the database management t e r m i n a l , see Figure 1). The s t r u c t u r e entry system c o n s i s t s of a PDP 11/04 computer w i t h 28K words of memory, a dual f l o p p y - d i s k d r i v e , keyboard, graphics t a b l e t , and CRT ( s i m i l a r to the DEC GT43 package). The graphics t a b l e t and a s s o c i a t e d s t y l u s enable a user to i n t e r a c t w i t h the d i s p l a y by moving the s t y l u s on the surface of the t a b l e t , r a t h e r than p o i n t i n g to the face of the scope as would be done w i t h a l i g h t pen. Software i n the computer t r a c k s the motion of the s t y l u s w i t h a cursor (a s m a l l cross) on the scope. Depressing the s t y l u s a c t i v a t e s a s w i t c h i n the s t y l u s t i p which i n t u r n allows the user to s e l e c t options from a "menu" on the d i s p l a y . Such a device has been found to be a very n a t u r a l medium f o r i n t e r a c t i n g w i t h a d i s p l a y and much more convenient than a l i g h t pen. A d d i t i o n a l det a i l s on the use of a graphics t a b l e t f o r chemical s t r u c t u r e drawing can be found i n references 7_ and 8_.


RETRIEVAL

110

O F MEDICINAL

CURRENT [ STRUCTURE I ι DATABASE,

CHEMICAL

INFORMATION

FUTURE f STRUCTURE I DATABASE,

HOST MACHINES

IBM 370/155 OS/VS1

COMMUNICATION CONTROLLER

FLOPPY DISK (DSD-210)

PROGRAM DEVEL OPMENT & STRUC TURE CORRECTION SYS. (PDP11/40)

GRAPHICS, KEYBOARD, TABLET

STRUCTURE ENTRY SYSTEM (PDP11/04)

PRINTER/ PLOTTER (VERSATEC)

FLOPPY DISK

GRAPHICS, KEYBOARD, TABLET

Figure 1. Hardware configuration for structure-entry project. High-volume structure entry was accomplished on the small graphics system (PDP 11/04); data base was formed on 370/155. Data-base maintenance and structure corrections were performed on the large graphics system (PDP 11/40). Information-retrieval-system-runs on 370/ 148 with data base transferred from 370/155.


8.

H O W E AND HAGADONE

Οπ-Line

Information

System

111

At times the host machine i s not a v a i l a b l e , and so r a t h e r than t r a n s m i t t i n g completed s t r u c t u r e s d i r e c t l y to the database on the 370, the program i n s t e a d w r i t e s them on a flopp y d i s k . These are l a t e r incorporated i n t o the database v i a the flopp y d i s k u n i t on the second graphics t e r m i n a l (see Figure 1). (b) I n t e r n a l S t r u c t u r e Representation. S t r u c t u r e s are r e presented i n the computer i n the form of atom-bond connection t a b l e s . These are arrays of data which account f o r such things as: ( i ) f o r each atom; atom type, formal charge, i s o t o p e l e v e l , presence of unpaired e l e c t r o n , two-dimensional c o o r d i nates, number of bonds attached, and a d d i t i o n a l i n f o r mation r e q u i r e d f o r regeneration of the s t r u c t u r e d i a gram, ( i i ) f o r each bond the bond, bon mation f o r cases where the bond i s attached to a c h i r a l atom. The connection t a b l e i s formed i n c r e m e n t a l l y during the s t r u c t u r e drawing o p e r a t i o n . Since X-Y coordinate data f o r each atom are stored i n the t a b l e , a complete molecular p i c t u r e can be generated almost i n s t a n t a n e o u s l y from the connection t a b l e . The t a b l e provides an unambiguous r e p r e s e n t a t i o n of a s t r u c t u r e ; how ever, a t the time the connection t a b l e i s i n s e r t e d i n the data base, a c a n o n i c a l i z a t i o n step (using a modified Morgan a l g o r i t h m (9j10)) i s performed which r e s u l t s i n a unique o r d e r i n g of the atoms w i t h i n the t a b l e and f a c i l i t a t e s a d i r e c t comparison of two " d u p l i c a t e " t a b l e s to detect d i f f e r e n c e s ( e r r o r s ) . The connection t a b l e that i s s t o r e d contains no h i g h e r - l e v e l chemical i n f o r m a t i o n such as a r o m a t i c i t y , r i n g i n f o r m a t i o n , o r stereochemical r e l a t i o n ships other than the bond type t r a n s c r i b e d from the hard copy r e cord. Such h i g h - l e v e l r e l a t i o n s h i p s (and others) can be e x t r a c t e d from the b a s i c i n f o r m a t i o n contained i n the t a b l e by a p p r o p r i a t e p e r c e p t i o n r o u t i n e s on the host machine. In f a c t , the s t r u c t u r e record that w i l l be used f o r high speed s u b s t r u c t u r e searching i s not the o r i g i n a l master connection t a b l e (CT) f o r each s t r u c t u r e , but a s p e c i a l l y formatted record derived from the CT which a l s o contains a l l the h i g h e r - l e v e l data necessary to provide search r e s u l t s i n as c l o s e to " i n t e r a c t i v e time" as p o s s i b l e (see d i s c u s s i o n of s u b s t r u c t u r e s e a r c h i n g ) . Thus, the c o n n e c t i v i t y informa t i o n w i l l a c t u a l l y be present i n more than one form i n the com p l e t e d r e g i s t r y and search system. In the f o l l o w i n g d i s c u s s i o n , however, "connection t a b l e " r e f e r s to the expanded c o n n e c t i v i t y array described at the s t a r t of t h i s s e c t i o n . (c) G r a p h i c a l S t r u c t u r e Input. S t r u c t u r e s are t r a n s c r i b e d i n t o the system from data sheets which c o n t a i n molecular formula, chemical name, s t r u c t u r e diagram, some p h y s i c a l and b i o l o g i c a l screening data, and a r e g i s t r y number c a l l e d a "U-number". I n


112

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

some r e s p e c t s , the g r a p h i c a l entry system i s s i m i l a r to those used by Chemical A b s t r a c t s S e r v i c e (11) f o r s t r u c t u r e input and by com p u t e r - a s s i s t e d s y n t h e s i s research groups (12 1Ζ 1) f o r s p e c i f i c a t i o n of t a r g e t molecules. There a r e , however, a number of d i f f e r ences from the l a t t e r systems due to our r e q u i r e d focus on e r r o r c o n t r o l , speed of e n t r y , and o v e r a l l s t r u c t u r e diagram cosmetics. A number of drawing options appear on the d i s p l a y , which es s e n t i a l l y represent a "menu" of graphics c o n t r o l s . To s e l e c t from the menu the user moves the s t y l u s on the t a b l e t so as to super impose the t r a c k i n g cursor on one of the o p t i o n s , and then de presses the s t y l u s s l i g h t l y to a c t i v a t e the d e s i r e d o p t i o n . As can be seen i n F i g u r e 2, at the top of the d i s p l a y a r e c t a n g l e appears around the TYPE o p t i o n . This i n d i c a t e s to the operator the o p t i o n that i s c u r r e n t l y a c t i v e . Some i n f o r m a t i o n must be entered v i a the keyboard This i n cludes the date and th s e s s i o n ) , and a U-numbe s t r u c t u r e i s drawn. The system matches the MF a g a i n s t the s t r u c ture when the OUTPUT o p t i o n i s s e l e c t e d and only t r a n s m i t s the s t r u c t u r e to the database i f the MF and s t r u c t u r e match. The l a r g e r e c t a n g l e i n the center of the d i s p l a y represents the drawing area i n s i d e which the molecular diagram i s c o n s t r u c t ed. E r r o r messages and other t e x t u a l feedback to the user appear at the bottom of the drawing area. The options which are arrayed along the top of the d i s p l a y a l l o w the user to change drawing modes. They operate as f o l l o w s . DRAW allows the user to perform a freehand drawing o p e r a t i o n to enter bonds and i m p l i c i t carbon atoms (see below f o r d e s c r i p t i o n ) ; RINGS changes the d i s p l a y to a second menu from which pre-drawn r i n g systems can be s e l e c t e d ; MOVE enables the user to a d j u s t the p o s i t i o n of atoms and t h e i r attached bonds by simply superimposing the cursor on the d e s i r e d atom and moving the s t y l u s (and thereby, the atom) to i t s new p o s i t i o n ; CENTER centers the drawing i n the box; DELETE a l l o w s the s e l e c t i v e erasure of atoms or bonds; TYPE r e t u r n s c o n t r o l to the keyboard; OUTPUT sends a completed s t r u c t u r e to the host machine a f t e r the molecule i s subjected to a s e r i e s of e r r o r checks (remaining e r r o r s are detected on the 370 during the d u p l i c a t e match); and CLEAN erases the drawing area and i n i t i a l i z e s the connection t a b l e . The three remaining options at the top of the d i s p l a y are f o r bond character m o d i f i c a t i o n . For example, the broken/zigzag l i n e allows s p e c i f i c a t i o n of stereochemical i n f o r m a t i o n . While the system i s i n t h i s mode, the user can " p o i n t to the center of a bond and i t w i l l become a dashed bond to i n d i c a t e a p r o j e c t i o n of the bond back i n t o the plane of the drawing. P o i n t i n g to the bond a second time converts i t to a "wavy" bond of the type normally used to i n d i c a t e undefined absolute c o n f i g u r a t i o n at a c h i r a l center. So f a r , t h i s has been s u f f i c i e n t to permit an adequate s p e c i f i c a t i o n of stereochemistry; however, i n the next v e r s i o n of the g r a p h i c a l e n t r y system wedge-shaped bonds w i l l be s p e c i f i a b l e , 3

9

11


8.

HOWE AND HAGADONE

On-Line

Information

System

113

to i n c r e a s e the f l e x i b i l i t y of the system and prevent any ambig u i t y of c h i r a l s i t e d e f i n i t i o n . The arrow o p t i o n above DELETE i s used f o r the s p e c i f i c a t i o n of s t r o n g l y p o l a r i z e d bonds where there i s a formal charge separat i o n between the ends of the bond. This can be used i n N-oxides or phosphates, f o r example. The r e g i s t r y and search system w i l l recognize the equivalence of R 3 N — > 0 and R 3 N 0 , so the a r row i s used mainly f o r cosmetic purposes without any l o s s o f s t r u c t u r a l i n f o r m a t i o n . And f i n a l l y , the s o l i d l i n e a t the top of the d i s p l a y i s used t o convert any of the s p e c i a l bond types j u s t described back t o a normal s i n g l e bond. Along the bottom of the d i s p l a y appear a number of commonlyo c c u r r i n g atom types and f u n c t i o n a l groups, as w e l l as some cont r o l o p t i o n s . The FLIP o p t i o n changes the bottom menu t o r e v e a l an a d d i t i o n a l s e t of l e s s commonly-occurring atoms and groups. F u n c t i o n a l groups not presen simply by s e l e c t i n g th n e c t i n g them w i t h the a p p r o p r i a t e bonds. T h i s , however, takes longer than i t does t o i n s e r t one of the predrawn groups. Many of the predrawn f u n c t i o n a l groups can a l s o be converted t o s t r u c t u r a l l y s i m i l a r groups. For example, t o draw a t r i c h l o r o m e t h y l group, the operator would (a) i n s e r t a C F 3 from the menu, (b) p i c k up a CI from the menu and superimpose i t on the F 3 i n the t r i f l u o r o m e t h y l group, and (c) depress the s t y l u s . This would immediately convert the C F 3 t o a C C I 3 . Since the C C I 3 i s r e p r e sented i n the connection t a b l e as three d i s t i n c t c h l o r i n e atoms attached t o the same carbon, the operator could a l s o enter the same group by drawing the three c h l o r i n e s s e p a r a t e l y . Although the appearance would be d i f f e r e n t , the c o n n e c t i v i t y data f o r the two forms would be the same. At the s t a r t o f each s t r u c t u r e drawing o p e r a t i o n , the comput e r requests the U-number and molecular formula of the compound. A f t e r t h i s i n f o r m a t i o n i s typed i n by the o p e r a t o r , the t a b l e t i s a c t i v a t e d and the p i c t u r e drawing stage can begin. Although there i s no drawing order imposed on the o p e r a t o r , the c y c l i c nucleus of the molecule i s u s u a l l y drawn f i r s t . Rings can be drawn i n two ways, freehand o r by s e l e c t i n g a predrawn r i n g system from the second d i s p l a y . To draw a bond "freehand", the user s e l e c t s the DRAW o p t i o n and then depresses the s t y l u s w i t h the cursor i n s i d e the drawing area. As the s t y l u s i s moved, a s t r a i g h t l i n e appears on the scope as i f i t were " i n k " from the s t y l u s . When the s t y l u s i s l i f t e d , the l i n e i s f r o z e n and the new bond i s i n s e r t e d i n the connection t a b l e . The t e r m i n a t i n g atoms are i n i t i a l l y assumed t o be carbon. A d d i t i o n a l bonds can be drawn i n t h i s manner u n t i l the d e s i r e d r i n g i s formed. We have found, however, that the entry o p e r a t i o n can be speeded up c o n s i d e r a b l y by p r o v i d i n g a c o l l e c t i o n of predrawn r i n g systems which can be brought i n t o view (Figure 3) by p r e s s i n g the RINGS o p t i o n . +


RETRIEVAL

Figure 2.

O F MEDICINAL

CHEMICAL

INFORMATION

Structure-entry dispfoy showing some of the graphics options avail able

DOUBLE

RETURN

U-59226A

CO OO Θ

Ο EE

Cc Figure 3.

A

P

Λ ^ o »

MOl/E

110 L .

Information

3

2

"

V-CH -N=CH-CH 2

0

\ = /

V

1 OH

STRUCTURE "

Κ

I.

F

Ν

C

- C a S i C l S H ?

L 1 HaB r Ρ

0

DOESN'T

MATCH M O L E C U L A R

h

+

C H

2

-*

+

C H

3

FLIP

FORMULA

COH^P

CH N H

OCH3

OH H N Ν sC

H 3 C CH3O

HO NH S O ^

2

C

2

Figure 6. Illustration of error detection prior to insertion of structure in data base. Message at bottom of drawing area says "structure doesnt match molecular formula."

USER 2

USER Ν

Figure 7. General hardware-component configuration of substructure-search sys tem. Front end consists of graphics minicomputers. Back end consists of dedicated minicomputer, "intelligent" disk controller, and dedicated disk.


RETRIEVAL

126

OF

MEDICINAL

CHEMICAL

INFORMATION

SS p o r t i o n s are e x t r a c t e d and sent to the SS minicomputer f o r exe c u t i o n . The SS minicomputer then sends the s u b s t r u c t u r e screen b i t s to the i n t e l l i g e n t d i s k c o n t r o l l e r and i n s t r u c t s i t to s t a r t scanning the d i s k . The s t r u c t u r e screens and connection t a b l e s w i l l be s t o r e d on the d i s k i n the format shown i n Figure 8. A ones-complement o p e r a t i o n i s performed on the s t r u c t u r e screen before i t i s w r i t t e n on the d i s k . Therefore, b i t s w i t h a value of 1 represent those s t r u c t u r a l a t t r i b u t e s that are absent i n the s t r u c t u r e . As the screen b i t s of each s t r u c t u r e pass the read head of the d i s k they are read by the c o n t r o l l e r and l o g i c a l l y AND ed w i t h che s u b s t r u c t u r e screen b i t s s u p p l i e d to the c o n t r o l l e r by the minicomputer. I f the r e s u l t of t h i s o p e r a t i o n i s nonzero the s t r u c t u r e cannot p o s s i b l y c o n t a i n the s u b s t r u c t u r e and i s e l i m i n a t e d from f u r t h e r c o n s i d e r a t i o n ; otherwise, the connect i o n t a b l e i s read i n t o the main memory of the minicomputer f o r f u r t h e r p r o c e s s i n g . Th when the end of a t r a c over to the next t r a c k . Scanning continues a f t e r a one r e v o l u t i o n delay. While the d i s k i s being scanned by the c o n t r o l l e r the minicomputer i s simultaneously executing the candidate s e l e c t i o n and atom-by-atom matching p o r t i o n s of the search. The atom and bond candidate s e l e c t i o n step i s performed by an a l g o r i t h m t h a t combines b i t screen and set r e d u c t i o n techniques. The connection t a b l e i s arranged i n a s p e c i a l format, w i t h one t a b l e entry f o r each bond i n the s t r u c t u r e . Each entry contains the atom types and sequence numbers of the atoms at each end of the bond as w e l l as the bond type. E n t r i e s are ordered by i n c r e a s i n g frequency of occurrence (based on s t a t i s t i c s c a l c u l a ted over the e n t i r e database) of the simple p a i r (atom-bond-atom sequence) c o n t a i n i n g the bond. In a d d i t i o n , a s m a l l number of screen b i t s , c a l l e d a p a i r screen, i s a s s o c i a t e d w i t h each bond. The p a i r screen, which i s a f u n c t i o n of atom and bond sequences w i t h i n a r a d i u s of 2 bond lengths of the c e n t r a l bond, d e s c r i b e s the s t r u c t u r a l environment i n the immediate neighborhood of the bond i n a manner s i m i l a r to that of a f u l l s t r u c t u r e screen. The p a i r screen b i t s are c a l c u l a t e d at the time the compound i s regi s t e r e d and are s t o r e d permanently i n the database. Although these e x t r a b i t s i n c r e a s e the s i z e of the database, experiments have shown t h a t they help provide short and r e l a t i v e l y c o n s i s t e n t search times. Execution proceeds by s e l e c t i n g , i n t u r n , each entry i n the s u b s t r u c t u r e t a b l e and screening against i t those e n t r i e s i n the s t r u c t u r e t a b l e t h a t are of the same simple p a i r type. The complemented screen b i t s of each q u a l i f y i n g s t r u c t u r e entry are l o g i c a l l y AND'ed w i t h the screen b i t s of the s u b s t r u c t u r e entry i n the same manner as f o r the f u l l s t r u c t u r e screen described above. A r e s u l t of zero i n d i c a t e s that the environment of the s e l e c t e d bond i n the s t r u c t u r e i s s i m i l a r to the environment of the current bond i n the s u b s t r u c t u r e . Candidate i n f o r m a t i o n i s s t o r e d for each s t r u c t u r e bond that matches the s u b s t r u c t u r e bond, to be f


8.

HOWE AND HAGADONE

Figure 8.

Οπ-Line

Information

System

Layout of structures on the disk for substruc ture searching


127

128

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

used into have ture

l a t e r i n the f i n a l atom-by-atom mapping of the s u b s t r u c t u r e the s t r u c t u r e . I f any s u b s t r u c t u r e atom or bond f a i l s to a candidate i n the s t r u c t u r e , the examination of t h a t s t r u c i s h a l t e d (a "no match" c o n d i t i o n ) . Figure 9 shows an example of the candidate s e l e c t i o n process u t i l i z i n g a s i m p l i f i e d p a i r screen of four b i t s per bond (although the optimal number of screen b i t s has yet to be determined, i t w i l l be i n the range of e i g h t to s i x t e e n b i t s per bond), which represent an adjacent s i n g l e bond, an adjacent double bond, an attached oxygen atom, and an attached carbon atom. In t h i s case, the f o l l o w i n g occurs: the t h i r d s t r u c t u r e entry i s screened a g a i n s t the f i r s t s u b s t r u c t u r e entry (same simple p a i r type) and passes the screen; the l a s t two s t r u c t u r e e n t r i e s are screened against the second s u b s t r u c t u r e entry and only the f o u r t h s t r u c t u r e entry passes the screen; and f i n a l l y , a l l of the s t r u c t u r e ent r i e s , except the t h i r d agains thir ture entry and only th i n d i c a t e the s t r u c t u r e bonds to which each s u b s t r u c t u r e bond has been mapped. I f a s t r u c t u r e passes the candidate s e l e c t i o n s t e p , an atomby-atom mapping of the s u b s t r u c t u r e i n t o the s t r u c t u r e i s performed and the r e g i s t r y numbers of compounds t h a t q u a l i f y are r e turned to the host machine as they are found. Since the SS system has yet to be implemented i n f i n a l form, accurate SS performance data are not a v a i l a b l e ; however, time p r o j e c t i o n s , based on current d i s k technology and an already implemented SS prototype system, i n d i c a t e that most searches w i l l r e q u i r e about 30 seconds elapsed time f o r the 60,000 compound database. 4.

I n t e g r a t i o n of B i o l o g i c a l Data:

Future

Goals

Although much work s t i l l needs to be done before the compound r e g i s t r y and search system w i l l be operated on a r o u t i n e b a s i s , most of the d i f f i c u l t problems concerning chemical s t r u c t u r e handling have been overcome. In the next major phase of the proj e c t the p r i n c i p a l e f f o r t w i l l focus on " b i o l o g i c a l data", a term which encompasses a very broad range of i n f o r m a t i o n i n the f i e l d of pharmacological s t u d i e s . The b i o l o g i c a l data h a n d l i n g capabil i t i e s of the query system w i l l undergo a c o n t i n u i n g e v o l u t i o n which w i l l come about not only as new types of pharmacological data become a v a i l a b l e f o r i n c o r p o r a t i o n i n t o the system, but a l s o as the need f o r (and a v a i l a b i l i t y o f ) new techniques f o r manipul a t i n g experimental data evolves. I n i t i a l work on the i n c o r p o r a t i o n of b i o l o g i c a l i n f o r m a t i o n i n t o the compound r e g i s t r y and search system w i l l d e a l mainly w i t h data that i s already being captured on a r o u t i n e b a s i s f o r computer input and storage. This i n c l u d e s screening r e s u l t s i n which the b i o l o g i c a l response of compounds to a v a r i e t y of t e s t screens i s i n d i c a t e d by numerical a c t i v i t y values or b i n a r y a c t i v i t y assignments ( a c t i v e / i n a c t i v e ) . A d d i t i o n a l data types to be i n c o r -



2

4

3

1 4

Figure 9.

2

5

6

=

_

STRUCTURE TABLE S M IR PLS F IC E IR D P A I EEN B O N D A T O M # 2 A T O M # 1 C ATOM #2TYPE TYPE TYPE - = 0 ATOM #1 1 4 C 0 1111 5 4 C 0 1 001 1 3 C 0 1011 1 2 C C 1110 5 6 C C 10 10

=

Simplified candidate-selection example (this is the second phase in a substructure search)

c-c-o-c-c

II

0

STRUCTURE

1

c-c-?

II

Ο

3

SUBSTRUCTURE

SUBSTRUCTURE TABLE ATOM #1BOND ATOM #2M IR F IC E IR D ATOM #1 ATOM #2TYPE TYPE TYPE PS A IPLS EEN 2 3 C 0 -1 0C 1 = 0O 1 2 C C 1110 ? 2 4 C 1111

130


porated w i l l e v e n t u a l l y i n c l u d e t o x i c i t y i n f o r m a t i o n and more det a i l e d a c t i v i t y r e s u l t s that p e r t a i n to i n d i v i d u a l c l a s s e s of pharmacological agents. As was mentioned i n the s e c t i o n on implementation, searches over the b i o l o g i c a l p o r t i o n of the database w i l l be c o n t r o l l e d by the r e l a t i o n a l database management system. The l o g i c c o n s t r u c t s of the expert query language w i l l a l l o w the user to s p e c i f y r a t h e r complex chemical and b i o l o g i c a l search requests i n which, for example, the database i s searched f o r a l l compounds that cont a i n a p a r t i c u l a r s u b s t r u c t u r e , which a l s o e x h i b i t a d e s i r e d a c t i v i t y l e v e l i n a given s c r e e n , and which a l s o were submitted a f t e r a p a r t i c u l a r date. Use of the RDBMS promises not only to reduce s u b s t a n t i a l l y the e f f o r t r e q u i r e d f o r i n t e g r a t i o n of the chemical and b i o l o g i c a l databases, but a l s o w i l l s i m p l i f y cons i d e r a b l y the e v o l u t i o n of b i o l o g i c a l l y - o r i e n t e d search c a p a b i l i ties. In a d d i t i o n to p r o v i d i n of chemical and b i o l o g i c a l data ( f o r d i s p l a y or r e p o r t generation purposes), an important feature of the system w i l l be i t s a b i l i t y to c r e a t e subsets of the main database. Users w i l l be able to t r e a t the r e s u l t s of t h e i r searches as t h e i r own p r i v a t e d a t a bases which can be accessed by s p e c i a l l y t a i l o r e d a p p l i c a t i o n programs. For example, compounds which were r e t r i e v e d by a combined s u b s t r u c t u r e and screening a c t i v i t y search could become source data f o r more d e t a i l e d analyses u s i n g p a t t e r n r e c o g n i t i o n , molecular m o d e l l i n g , or s t a t i s t i c a l techniques. Although we expect that the b i o l o g i c a l i n f o r m a t i o n h a n d l i n g c a p a b i l i t i e s of the system w i l l undergo a c o n t i n u i n g e v o l u t i o n , there i s a need f o r the i n c l u s i o n of other types of data as w e l l . S p e c t r a l data, patent s t a t u s i n f o r m a t i o n , CAS r e g i s t r y numbers, chemical names, and p h y s i c a l property data a l l f a l l under the umb r e l l a of " m e d i c i n a l chemical i n f o r m a t i o n " and are some of the more important data types that have been planned f o r eventual i n c l u s i o n i n the system. The p r o j e c t e d c a p a b i l i t i e s of the system, enabling a user to i n t e r a c t i v e l y query and manipulate such d i v e r s e types of i n f o r m a t i o n , should make the system an important asset i n the research and research management f u n c t i o n s . Literature Cited 1.

2. 3. 4.

Computer-Assisted Organic Synthesis, Wipke, W. T. and Howe, W. J., e d s . , ACS Symposium S e r i e s No. 61, American Chemical S o c i e t y , Washington, D . C . (1977). Minicomputers and Large Scale Computations, Lykos, P., e d . , ACS Symposium S e r i e s No. 57 (1977). Computer-Assisted Structure Elucidation, Smith, D. H., e d . , ACS Symposium S e r i e s No. 54 (1977). Chemometrics: Theory and Application, K o w a l s k i , B. R . , e d . , ACS Symposium S e r i e s No. 52 (1977).


8.

HOWE AND HAGADONE

On-Line Information System

5.

131

Algorithms for Chemical Computations, C h r i s t o f f e r s o n , R. E., ed., ACS Symposium S e r i e s No. 46 (1977). 6. Computer Networking and Chemistry, Lykos, P., ed., ACS Sym posium S e r i e s No. 19 (1975). 7. Howe, W. J. and Hagadone, T. R., "Substructure S e a r c h i n g " , in Proceedings of the Technical Information Retrieval Com mittee of the Manufacturing Chemists Association, Washington Meeting, 1977, in p r e s s . 8. Corey, E . J. and Wipke, W. T., Science, 166, 178 (1969). 9. Morgan, H . L., J. Chem. Doc., 5, 107 (1965). 10. Wipke, W. T. and D y o t t , T. M., J. Amer. Chem. Soc., 96, 4834 (1974). 11. B l a k e , J. Ε., Farmer, Ν. Α . , and Haines, R. C., J. Chem. Inf. and Computer Sci., 17, 223 (1977). 12. Corey, E . J., Wipke, W. T., Cramer, R. D., and Howe, W. J., J. Amer. Chem. Soc, 13. Wipke, W. T., in Computer Chemical Information, Wipke, W. T., Heller, S. R., Feldman, R. J., Hyde, E., e d s . , p . 147, Wiley Publ., New York (1974). 14. Brown, H. D., Castlow, Μ., C u t l e r , Ε. Α . , Jr., Demott, Α. Ν., Gall, W. B., Jacobus, D. P., and Miller, C. J., J. Chem. Inf. and Computer Sci., 16, 5 (1976). 15. Codd, E . F., "A Relational Model of Data f o r Large Shared Data Banks", Commun, of the ACM, XIII, 377 (1970). 16. Codd, E . F., "Further N o r m a l i z a t i o n of the Data Base R e l a tional Model", Courant Computer Science Symposia 6, Data Base Systems, Prentice-Hall, New York (1971). 17. Date, C. J., An Introduction to database Systems, Addison Wesley, Reading, M a s s . , (1975). 18. A s t r a h a n , M. M., et al, "System R. R e l a t i o n a l Approach to Database Management", A.C.M. Transactions on Database Sys tems, 1, 97 (1976). 19. Feldman, Α . , Hodes, L., J. Chem. Doc., 15, 147 (1975). 20. Adamson, G. W., Bush, J. Α . , M c l u r e , Α . , and Lynch, M. F., J. Chem. Doc., 14, 44 (1974). 21. Meyer, Ε., "Superimposed Screens f o r the GREMAS System", i n Proc. FID-IFIP Conference, p . 280, Samuelson, Κ., ed., Rome Meeting, 1967, North Holland P u b l . (1968). 22. Sussenguth, Ε. H., Jr., J. Chem. Doc., 5, 36 (1965). 23. F i g u e r a s , J., J. Chem. Doc., 12, 237 (1972). 24. H a i n e s , R. C., "Substructure Search Design Study Status Re p o r t " , Chemical A b s t r a c t s S e r v i c e Working Paper (unpublished), 1976. 25. Bird, R. M., Tu, J. C., Worthy, R. Μ., "Associative/Parallel Processors for Searching Very Large Textual Data Bases", SIGIR-SIGARCH-SIGMOD T h i r d Workshop on Computer A r c h i t e c t u r e for Non-numeric P r o c e s s i n g , McGill, M. J., ed., SIGMOD, 9, No. 2, 8 (1977). RECEIVED August 29, 1978.


9

Warner-Lambert/Parke-Davis-CAS

Registry III

Integrated I n f o r m a t i o n System ROGER D. WESTLAND, RAYMOND L. HOLCOMB, JOHN W. VINSON, JON D. STEELE, ROBERT J. CARDWELL, ROBERT L. SCOTT, THOMAS D. HARKAWAY, PATRICIA J. HYTTINEN, and TINA WILLIAMS Warner-Lambert/Parke-Davis Pharmaceutical Research Division, Ann Arbor, MI 48105 In 1946 the Parke-Davi h Laboratories c e n t r a l i z e d chemical and biological storage and retrieval. These were e f f e c t i v e until the l a t e 1950's, when manual methods were g r a d u a l l y r e i n f o r c e d w i t h punched card files. By the mid 1960's, machine readable data files were a v a i l a b l e f o r everything except a complete chemical s t r u c t u r e and certain other s t r u c t u r e - r e l a t e d i n f o r m a t i o n . Throughout the development o f computerized i n f o r m a t i o n systems it has been necessary to m a i n t a i n redundant manual files until n e a r l y all i n f o r m a t i o n is computer-readable. Only now, a f t e r adding chemical s t r u c t u r e s to the computer database can we b e g i n to abandon the manual files maintained f o r over 30 y e a r s . I n a d d i t i o n to s t r u c t u r e - h a n d l i n g capability, we have developed a system to link sample i n v e n t o r y and p r o p e r t i e s , b i o l o g i c a l screening d a t a , and research document data to produce reports and answers to q u e r i e s , both interactively and in batch mode. In c o n s i d e r i n g approaches to computerized chemical s t r u c t u r e p r o c e s s i n g (I, 2, 3), we accepted an o f f e r by Chemical A b s t r a c t s S e r v i c e (CAS) to e s t a b l i s h under c o n t r a c t a p r i v a t e s a t e l l i t e o f the CAS R e g i s t r y System (4) which employs over 640 programming modules and over a q u a r t e r - m i l l i o n source statements. Since Warner-Lambert/Parke-Davis (WL/PD) had compatible hardware f o r both p r o c e s s i n g and s t r u c t u r e p r i n t i n g , we were i n a p o s i t i o n to take advantage o f CAS s l a r g e investment i n h i g h q u a l i t y graphics, name p r o c e s s i n g , and computer e d i t s . CAS offered an advanced and h i g h l y developed system which could be i n s t a l l e d i n a short time at r e l a t i v e l y low c o s t . Ongoing development at CAS to enhance the system f o r s t o r i n g , r e t r i e v i n g , and r e p o r t i n g the w o r l d ' s chemical l i t e r a t u r e made c o m p a t i b i l i t y w i t h CAS a t t r a c t i v e . Current use o f CAS s s e r v i c e i n Europe (j>) , Japan ( 6 ) , and the United States (7., 8, 9) evidences i n c r e a s i n g r e l i a n c e on the CAS R e g i s t r y System and suggests the p o s s i b i l i t y o f broad i n d u s t r i a l and governmental use i n the f u t u r e . A p i l o t p r o j e c t at WL/PD r e q u i r e d l e s s than two-man months of e f f o r t to implement CAS*s s t r u c t u r e - p r i n t i n g algorithms from 1

1

0-8412-0465-9/78/47-084-132$05.00 Published 1978 American Chemical Society In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

9.

WESTLAND

E TAL.

Integrated

Information

System

133

the CAS G r a p h i c a l Data S t r u c t u r e (10, JL1, 12) r e c o r d . Success o f the experiment i n p l o t t i n g s t r u c t u r e s o f the type shown i n F i g u r e 1 s t i m u l a t e d f u r t h e r e x p l o r a t i o n which u l t i m a t e l y l e d to the development o f a WL/PD - CAS i n t e g r a t e d system f o r s t o r i n g , r e t r i e v i n g , manipulating, and r e p o r t i n g chemical and b i o l o g i c a l research data. System Design With an INQUIRE® (Infodata Systems Inc., F a l l s Church, V i r g i n i a ) database management system a v a i l a b l e on our IBM 370/168 computer, h i s t o r i c a l computer f i l e s o f sample i n v e n t o r y and t r a n s a c t i o n s , p h y s i c a l and chemical p r o p e r t i e s , b i o l o g i c a l screening data, research document data, and other miscellaneous f i l e s were converted to INQUIRE f i l e format (13), and stored on d i s k (Figure 2 ) . Sampl storage) and i n v e n t o r y c e n t r a l computer an o n - l i n e balance and a keyboard-CRT t e r m i n a l . Other WL/PD i n f o r m a t i o n i s entered i n a k e y - t o - d i s k o p e r a t i o n using the ENTREX® (14) system, thereby p r o v i d i n g options f o r d i r e c t entry o f data from l a b o r a t o r i e s , when a p p r o p r i a t e . Output from the P r i v a t e R e g i s t r y f i l e s a t CAS i s converted by means o f update programs t o INQUIRE f i l e formats. M u l t i - f i l e searching o f the INQUIRE f i l e s f o r ad hoc queries o r r e p o r t c o n s t r u c t i o n can be done e i t h e r i n t e r a c t i v e l y w i t h TSO terminals o r i n batch mode using a V a r i a n V74 computer as a HASP w o r k - s t a t i o n . Generic s t r u c t u r e searches o f the computer f i l e o f fragment-coded s t r u c t u r e s g i v e as o p t i o n a l output punched paper tape that cont r o l s the d i s p l a y o f s t r u c t u r e images on m i c r o f i c h e . The coded m i c r o f i c h e c o n t a i n i n g 196 s t r u c t u r e images a t 24X r e d u c t i o n are stored i n the c a r o u s e l o f a storage and r e t r i e v a l u n i t manufactured by Image Systems, Inc. Since a new s u b s t r u c t u r e search system f o r the WL/PD f i l e w i l l not be usable u n t i l the e n t i r e backlog o f s t r u c t u r e s has been entered i n t o the P r i v a t e R e g i s t r y , we are cons i d e r i n g programs to a l g o r i t h m i c a l l y generate the Parke-Davis Fragmentation Code (15) from CAS connection t a b l e s . This w i l l a l l o w us to continue using our present search techniques i n the interim. P r o p e r t i e s F i l e . The f o l l o w i n g data are i n c l u d e d i n the keyt o - d i s k entry o f p r o p e r t i e s : a c c e s s i o n number, source, percent of parent component, m e l t i n g o r b o i l i n g p o i n t , s p e c i a l handling or storage requirements, p h y s i c a l s t a t e , s o l u b i l i t y , s t a b i l i t y , s e l e c t e d a n a l y t i c a l and s p e c t r a l data, sample weight and l o c a t i o n , submission date, and l i t e r a t u r e r e f e r e n c e s . Transactions F i l e . A M e t t l e r Model PT320 balance having BCD output, and a CRT t e r m i n a l are i n t e r f a c e d w i t h the c e n t r a l computer through a microprocessor and the V a r i a n HASP w o r k - s t a t i o n . At the time sample weights are a u t o m a t i c a l l y recorded, the


RETRIEVAL

134

O F MEDICINAL

CHEMICAL

INFORMATION

operator keys 1) t r a n s a c t i o n type, 2) a c c e s s i o n number, 3) date, 4) whether the sample i s being r e c e i v e d and from whom, o r being t r a n s m i t t e d and to whom, and 5) storage l o c a t i o n . While t h i s i n f o r m a t i o n i s s t o r e d i n the " T r a n s a c t i o n s " database a running record o f the amount o f sample on hand i s c a l c u l a t e d from onl i n e balance e n t r i e s and s t o r e d i n the " P r o p e r t i e s " database. B i o l o g y F i l e . Screening data from b i o l o g y l a b o r a t o r i e s a r e recorded on data e n t r y forms a p p r o p r i a t e l y coded (13) f o r key-tod i s k h a n d l i n g , e i t h e r i n a c e n t r a l l o c a t i o n o r the l a b o r a t o r y itself. R e s u l t forms are customized f o r each t e s t and a r e rearranged i n t o a standard format by the ENTREX processor before being sent to the main computer. Document F i l e . Search parameters o f i n t e r n a l l y generated r e s e a r c h r e p o r t s are i n c l u d e (word) p r o c e s s i n g equipment inexpensive r e c o r d i n g o f s e l e c t e d t e x t such as a b s t r a c t s . A v a r i e t y o f o p t i o n s to INQUIRE i n c l u d e techniques which can index and r e t r i e v e on the b a s i s o f such t e x t . The m u l t i - f i l e o p t i o n allows s e l e c t e d records t o be combined w i t h data from other INQUIRE f i l e s . CAS F i l e s . Machine p r o c e s s i n g o f data must be performed a t CAS to take advantage o f the many machine v a l i d a t i n g and d u p l i cate checking f e a t u r e s o f the CAS R e g i s t r y System. Although s t r u c t u r e s and chemical names could be entered a t the user's l o c a t i o n followed by t r a n s m i t t a l o f computer-readable data t o CAS f o r p r o c e s s i n g , CAS s keyboarding conventions and h i g h volume a l l o w them to o f f e r the s e r v i c e a t a cheaper r a t e than we could match i n t e r n a l l y . A c c o r d i n g l y , data sheets o f chemical s t r u c t u r e s and names are shipped to CAS on a twice-weekly b a s i s ( F i g u r e 3). A t CAS the hand-written i n f o r m a t i o n i s checked and e d i t e d , and s t r u c t u r e s , s t e r e o - d e s c r i p t o r s , and names are entered by a k e y - t o - d i s k procedure (17) . Keyboarded records o f s t r u c tures are processed i n the P r i v a t e R e g i s t r y s a t e l l i t e system w i t h the use o f most o f the computer e d i t s o f the CAS R e g i s t r y System (17). A d i s t i n g u i s h i n g f e a t u r e o f t h i s process i s a check t o determine i f the newly entered s t r u c t u r e a l s o e x i s t s i n the R e g i s t r y f i l e o f over four m i l l i o n substances. I f an exact d u p l i cate i s found i n the CAS f i l e , the CAS R e g i s t r y Number along w i t h the CA Index name and synonyms are returned as an update t o the WL/PD Names F i l e . C r i t i c a l to the d u p l i c a t e check as c u r r e n t l y handled i s that the e n t i r e s t r u c t u r e , i n c l u d i n g the s a l t or s o l vate p o r t i o n , must be i d e n t i c a l even as to the p r o p o r t i o n o f components o f a multi-component s t r u c t u r e (e.g., RNI^'I^SC^ does not match RNH2*1/2H2S04. System m o d i f i c a t i o n s could remove the l i m i t a t i o n . A p r o f i l e o f a l l WL/PD substances entered i n t o the p r i v a t e WL/PD system i s maintained by CAS and checked p e r i o d i c a l l y f o r matches i n the CAS R e g i s t r y f i l e s . Therefore, w i t h i n 1


WESTLAND E T AL.

Integrated Information System

OH

Figure 1. Plotted structure of chalcomycin (CAS Registry Number 20283-48-1). Stereochemistry is provided by a "text descriptor" which is printed along with the topological representation shown. DOCUMENTS

BIOLOGY

BIOLOGY

INVENTORY INQUIRE DATABASES

DATA ENTRY

BIODATA

PROPERTIES f

CONVERSION

M r r

X

RETRIEVAL

O F MEDICINAL CHEMICAL INFORMATION

NUMBER OF COMPOUNDS

FILE NIH/EPA-MSSS C-13 N M R

25,560 3,765

E P A - A C T I V E I N G R E D I E N T S IN P E S T I C I D E S PESTICIDES S T A N D A R D S ORD —CHEMICA OIL A N D H A Z A R D O U AEROS/SAROAD AEROS/SOTDAT STORET C H E M I C A L SPILLS T S C A I N V E N T O R Y C A N D I D A T E LIST NIMH-PSYCHOTROPIC

DRUGS

S R I - P H S LIST 149 O F C A R C I N O G E N S N B S — S I N G L E C R Y S T A L FILE H E A T S O F F O R M A T I O N O F G A S E O U S IONS G A S - P H A S E P R O T O N AFFINITIES N S F - R A N N P O L L U T A N T FILE FDA-PESTICIDE

REFERENCE STANDARDS

CPSC-CHEMRIC

MONOGRAPHS

CAMBRIDGE

UNIVERSITY

CRYSTAL DATA

1,454 384

65 572 234 577 33,579 1,686 4,448 18,362 3,169 454 225 613 1,000 10,018

EROICA T H E R M O D Y N A M I C D A T A

4,492

M E R C K INDEX

8,894

ITC — I N T E R N A T I O N A L T R A D E C O M M I S S I O N

9,194

N I O S H - R E G I S T R Y OF TOXIC EFFECTS OF CHEMICAL SUBSTANCES N F P A - H A Z A R D O U S CHEMICALS

19,908 397

Figure 2. List of the current 25 collections which currently comprise the CIS unified data base (integrated SANSS data base 3/1/78)


10.

HELLER AND MILNE

NIH/EPA

Chemical Information System

147

Among the data bases being added t o the CIS t h i s year are those shown i n Figure 3. Over the next 2-3 years, w i t h the cont i n u e d a d d i t i o n o f f i l e s that are e i t h e r generated or used by the Government, i t i s expected that the l i s t of r e f e r r a l f i l e s w i l l grow to over 250. With the recent e f f o r t s of the four main F e d e r a l r e g u l a t o r y Agencies (EPA, FDA, CPSC, OSHA) to coordinate t h e i r v a r i o u s a c t i v i t i e s , such as the study and r e g u l a t i o n of s p e c i f i c chemicals, t h i s c e n t r a l r e f e r r a l system takes on more importance. This four-Agency group, known as the Interagency Regulatory Liason Group (IRLG) ( 3 ) , i s now working to use the Chemical A b s t r a c t s Servic e (CAS) R e g i s t r y Number as the standard chemical i d e n t i f i e r f o r the chemicals i n a l l the four Agencies. An i n t e r n a l r e g u l a t i o n has been proposed which w i l l make t h i s mandatory. The r e g u l a t i o n i s modelled a f t e r EPA Order 2800.2, c u r r e n t l y the o n l y Government r e g u l a t i o n to mandate standardized chemical c l a s s i f i c a t i o Over the past fou been submitted t o CAS, under c o n t r a c t t o EPA, t o o b t a i n the CAS R e g i s t r y Numbers f o r these chemicals. The r e s u l t o f t h i s massive and c o s t l y e f f o r t i s the CIS U n i f i e d Data Base (UDB) of about 101,000 unique chemicals a s s o c i a t e d w i t h the 25 f i l e s shown i n F i g u r e 2. That there i s so much overlap of the chemicals found i n these f i l e s i s not s u r p r i s i n g . I t i s beginning to appear that there are r e l a t i v e l y few chemicals which are a c t u a l l y studied i n any d e t a i l , and even fewer that become s i g n i f i c a n t i n commerce, as, f o r example, drugs, food a d d i t i v e s or p e s t i c i d e s . P r o j e c t i o n s suggest that by the time the CAS r e g i s t r a t i o n process o f some 250 f i l e s i s completed, the a c t u a l s i z e o f the CIS u n i f i e d Data Base w i l l not exceed 175,000-200,000 substances. The need then w i l l be to o b t a i n as much u s e f u l and accurate i n f o r m a t i o n about these substances as i s necessary to p r o t e c t h e a l t h and environment i n the USA, as i s r e q u i r e d by the missions o f our r e s p e c t i v e Agencies. I t i s our hope that by d e f i n i n g the s i z e o r scope of the " r e a l " universe o f chemicals, that the burden on i n d u s t r y w i l l be lessened and that f u t u r e e f f o r t s w i l l be e a s i e r to d i r e c t . Thus, we see l i t t l e immediate need t o study the universe that CAS has d e f i n e d , of over some 4,000,000 chemicals found i n the l i t e r a t u r e that CAS has a b s t r a c t e d s i n c e 1965. Only about 12% of these four m i l l i o n have appeared more than once i n the CASa b s t r a c t e d l i t e r a t u r e and probably no more than 3% are produced and s o l d i n anything but research q u a n t i t i e s . S t r u c t u r e and Nomenclature Search System (SANSS) The S t r u c t u r e and Nomenclature Search System (SANSS), the heart o f the CIS, i s based upon the work of Feldraann who developed the o r i g i n a l search algorithms a number o f years ago (5). A d d i t i o n of a nomenclature search program, an i d e n t i t y search program and a search program based on the Edgewood CIDS s t r u c t u r e keys ( 6 ) , as w e l l as some c o n s i d e r a b l e refinement of the system

American Chemical Society Library 1155 16th St. N. W. In Retrieval ofWashington, Medicinal Chemical Information; D. C. 20036Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

RETRIEVAL OF MEDICINAL C H E M I C A L INFORMATION

U.S. Coastguard Chemical Properties File. E P A I E R L Non-Criteria Pollutant Emissions. E P A , Section 111A of the Clean A i r Act. E P A , Office of Air Quality, Permissible Standards, Criteria Pollutants. E P A , Office of Water Supply File of Drinking Water Pollutants. E P A , Pollutant Strategies Branch, Selected Organic A i r Pollutants. E P A , Effluent Guidelines Consent Decree List E P A , Section 112 of the Clean Air Act. E P A , O R D , Gulf Breeze, List of Chemicals. E P A , Carcinogen Assessment G r o u p List of Chemicals.

E P A , List of Potentially Hazardous Chemicals f r o m Coal and O i l . California O S H A List of Chemical Contaminants. W H O , F o o d and Agriculture Organization, List of Pesticides. E P A , I E R L , Organic Chemicals in Air

Carcinogens. N C T R , Potential Industrial Carcinogens and Mutagens. E P A , I E R L , List of Environmental Carcinogens. E P A , OPP, Pesticide Literature Searches. N I E H S , Laboratory Chemicals. T o x i c and Hazardous Industrial Chemicals Safety Manual. International Technical Information Institute, T o k y o .

E P A , R P A R Candidates Chemical Review Schedule List.

List of Teratogenic Chemicals. Medical Information Center, Karolinska Institute, Stockholm.

E P A , O T S Status Assessments.

E P A , List of Hazardous Pesticides.

E P A , Standing Air Monitoring Work Group List of N o n Criteria Pollutants.

E P A , Mutagenicity Studies. C I T T , List of Candidates.

EPA, O R D - O H E E Chemicals.

E P A , T S C A Section 8e, List of Chemicals.

Figure 3.

Laboratory

Newfilesbeing added to the NIH/EPA CIS UDB in Spring, 1978


10.

HELLER AND M I L N E

NIH/EPA

Chemical Information System

149

has been c a r r i e d out over the l a s t few years. The SANSS and i t s data base, connection t a b l e s from CAS and chemical names, has absorbed the bulk o f the CIS budget. C u r r e n t l y , the SANSS can be used i n a number o f ways. The more important methods a r e : * Nomenclature Search (NPROBE) * Ring Search (RPROBE) * Fragment search (FPROBE) * CIDS code search (SPROBE) * Molecular weight search (MW) * Molecular formula search (MF) * Substructure search (SUBSS) * F u l l s t r u c t u r e search (IDENT) In a d d i t i o n t o these searching programs, there are a number o f r e t r i e v a l and d i s p l a y o p t i o n s a v a i l a b l e i n the system. These include: * Displa * D i s p l a y of CAS C o l l e c t i v e Index names * D i s p l a y of synonyms, common names and trade names * D i s p l a y of molecular formulas * D i s p l a y of f i l e s c o n t a i n i n g a substance * R e t r i e v a l based upon CAS R e g i s t r y Number The f o l l o w i n g s e c t i o n s w i l l be devoted to e x p l a i n i n g the v a r i o u s SANSS nodules and g i v i n g examples o f how they can be used. At the end o f the chapter an example o f the i n t e r f a c i n g of the SANSS w i t h the NIOSH RTECS data base o f acute t o x i c i t y data (7) w i l l be des c r i b e d , as an example of the d i r e c t i o n that CIS development i s t a k i n g . Since there i s c o n s i d e r a b l e i n t e r e s t on the part o f the chemical i n d u s t r y i n the implementation o f TSCA, access t o the b u l k of the p u b l i c data t h a t EPA w i l l be u s i n g i n i t s work f o r a d m i n i s t e r i n g TSCA should be of v a l u e . At present, development of the SANSS i s being d i r e c t e d towards the immediate needs of EPA s O f f i c e o f Toxic Substance (OTS), so that the foundation t h a t has been b u i l t f o r the SANSS can be used most e f f e c t i v e l y f o r the implementation of TSCA. 1

Name - Nomenclature Search (NPROBE) The name search, NPROBE, has been implemented as a r e s u l t of requests expressed by both the SANSS user community and the CEQTSCA MITRE study proposal (8) f o r the development o f a Chemical S t r u c t u r e and Nomenclature System which we have c a l l e d the S t r u c t u r e and Nomenclature Search System. The software used i s s i m i l a r t o t h a t used i n the ÇHEMLINE system a t the N a t i o n a l L i b r a r y o f Medicine (NLM) and a l l o w s f o r complete or p a r t i a l (fragment) name search. There are an average o f s l i g h t l y over 3 names per chemical i n CIS UDB, as opposed t o s l i g h t l y more than 2 names per chemical i n CHEMLINE ( 9 ) . The CHEMLINE f i l e , which l i n k s p r i m a r i l y t o the TOXLINE l i t e r a t u r e r e f e r e n c e s , i s made up


150

RETRIEVAL

OF MEDICINAL CHEMICAL

INFORMATION

mostly of r e s e a r c h chemicals, and thus i s not l i k e l y to have the m u l t i p l e synonyms t h a t are a s s o c i a t e d w i t h commercial chemicals. In the CIS UDB, which i s comprised of f i l e s from p r i m a r i l y regu l a t o r y , and hence commercial, sources, there are the expected a d d i t i o n a l names a s s o c i a t e d w i t h m a t e r i a l s i n commerce. To conduct a nomenclature search, the user simply enters a chemical name or name fragment, as shown i n F i g u r e 4. The example shown i n F i g u r e 4 i s of a search f o r any substance i n the UDB whose name contains the fragment "DDT . From F i g u r e 4 i t can be seen that there are 12 such substances i n the UDB, of which the f i r s t , p,p DDT, i s shown i n the F i g u r e . In a d d i t i o n , a l s o shown i n t h i s f i g u r e a r e a l l the f i l e s of the UDB which c o n t a i n i n f o r mation on p,p DDT, w i t h the l o c a l f i l e i d e n t i f i e r numbers l i s t e d so that one may go d i r e c t l y to the p a r t i c u l a r f i l e and get the i n f o r m a t i o n t h a t i s contained i n that f i l e regarding ρ,ρ' DDT. I n F i g u r e 5, a name searc on the e n t i r e UDB and f i v these f i v e i s shown i n F i g u r e 5, w i t h the names of the f i l e s that have i n f o r m a t i o n about LSD. Not s u r p r i s i n g l y , the f i l e s i n c l u d e the NIMH L i s t of Psychotropic Drugs, the Merck Index and the NIOSH acute t o x i c i t y data base, as w e l l as the NIH/EPA Mass S p e c t r a l Data Base and the TSCA Candidate L i s t . There i s l i t t l e doubt t h a t the i n c l u s i o n on the TSCA Candidate or "Strawman * l i s t w i l l be changed once the f i n a l TSCA i n v e n t o r y i s p u b l i s h e d , s i n c e under present law, LSD i s an i l l e g a l chemical substance. This i s a use f u l search technique, but r e q u i r e s a l a r g e l i s t o f synonyms, a c o r r e c t s p e l l i n g , and a knowledge of how chemical names a r e broken down. For example, i n searching f o r a cyclohexanedione, i f the f i l e name o f the substance i s w r i t t e n as 2,5-cyclohexanedione r a t h e r than cyclohexan-2,5-dione, a search f o r "dione w i l l not f i n d the chemical. 11

T

f

1

11

F u n c t i o n a l Group - CIDS Key Search (SPROB) The best way t o search f o r f u n c t i o n a l groups o r s t r u c t u r e f e a t u r e s i n the CIS SANSS i s to use the Chemical Information Data Systems (CIDS) keys, developed by Edgewood A r s e n a l . The CIDS keys, a few o f which are shown i n F i g u r e 6, are the b a s i s o f a r a p i d and e f f i c i e n t way to search the CIS UDB f o r substances c o n t a i n i n g a p a r t i c u l a r f u n c t i o n a l group or s t r u c t u r e f e a t u r e . Many of the CIDS keys are q u i t e s p e c i f i c i n nature, as can be seen i n F i g u r e 6. Others, shown towards the bottom of F i g u r e 6, a r e q u i t e generic i n nature. For example, the CIDS key FG25 r e f e r s to the presence of a n i t r i l e or cyanide group i n the molecule. An example o f a CIDS key search i s given i n F i g u r e 7, where a search i s shown f o r a l l c y c l o h e x y l (SCN49) morpholine (SCN35) compounds i n the NIOSH RTECS data base of acute t o x i c i t y . There are o n l y two such compounds i n the data base, and the f i r s t o f these i s p r i n t e d out i n the f i g u r e , along w i t h i t s l o c a l NIOSH RTECS i d e n t i f i e r numbers i n d i c a t e d .


HELLER

AND MILNE

OPTION?

NIH/EPA

Chemical

Information

System

NPROBE

FRAGMENT OR WHOLE NAME SEARCH (F/W) (F) ? F S P E C I F Y F R A G M E N T (CR T O EXIT): D D T FILE 1, 12 C O M P O U N D S H A V I N G F R A G M E N T : D D T S P E C I F Y F R A G M E N T (CR T O EXIT): _ OPTION? SSHOW 1 HOW M A N Y S T R U C T U R E S (E T O E X I T ) ? 1_ TYPE Ε T O TERMINATE DISPLAY STRUCTURE 1 C A S R E G I S T R Y N U M B E R 50-29-3 T S C A C A N D I D A T E LIST: R000-2373 CIS M A S S S P E C T R O M E T R Y CIS C A R B O N 13 N M R S P E C T R O M E T R Y :

50-29-3.01

EPA PESTICIDES - A C T I V E INGREDIENTS: EPA OHM/TADS:

CAMBRIDGE XRAY CRYSTAL: MERCK

29201

72T16510 50-29-3.01

INDEX

EPA PESTICIDES - A N A L Y T I C A E P A S T O R E T : 39317, 39373, 39371,39374, 39372, 39370, 39359, 39375 39376, 39378, 39290, 39358, 39377, 39302, 39303, 39304, 39300, 39301 EPA C H E M I C A L SPILLS CPCSCHEMRIC F D A / E P A PESTICIDES R E F . S T A N D A R D S : 200 U.S. I N T E R N A T I O N A L T R A D E C O M M I S S I O N N B S X R A Y C R Y S T A L : 50-29-3.01 N S F C H E M I C A L S LIST: 138 PHS-149 C A R C I N O G E N S : A0240 NIOSH RTECS: KJ33250 C14H9C15

C

CL*C

C

C

C

C

C**CL

C** **C**C

C

#

C

CL*C**CL

C

CL Benzene, 1, 1, — ( 2 , 2, 2-trichloroethylidene) bis [ 4 - c h l o r o - (9CI) Ethane, 1, 1, 1-trichloro-2, 2-bis (p-chlorophenyl)— (8CI) .alpha., .alpha. —Bis (p-chlorophenyl)—.beta., .beta., .beta, -trichlorethane p, p' -Dichlorodiphenyltrichloroethane p, p' - D D T

Figure 4.

NPROBE

name search for name fragment "DDT*


RETRIEVAL

O F MEDICINAL

OPTION? N P R O B E F R A G M E N T O R W H O L E N A M E S E A R C H (F/W) (F) ?F S P E C I F Y F R A G M E N T (CR T O E X I T ) : LSD FILE 5, 5 COMPOUNDS HAVING FRAGMENT: S P E C I F Y F R A G M E N T (CR T O E X I T ) : _ O P T I O N ? SSHOW 5 HOW M A N Y S T R U C T U R E S (E T O EXIT) ? 1 TYPE Ε TO TERMINATE DISPLAY STRUCTURE 1 C A S R E G I S T R Y N U M B E R 50-37-3 T S C A C A N D I D A T E LIST: R000-3157 CIS M A S S S P E C T R O M E T R Y MERCK INDEX NIMH P S Y C H O T R O P I NIOSH R T E C S : KE42000,KE41000,KE4375

CHEMICAL

INFORMATION

LSD

C20H25N30

C. .C

N**C

*

c. .c

* C++C

C++C •

*

•

•

*

*

C**C

ο

c

+

•

+

•

C**C *N**C**C #

N**C

Ergoline-8-carboxamide, 9, 10-dtdehydro-N, N-diethyl-6-methyl—, (8.beta.) (9CI) Ergoline-8 .beta, -carboxamide, 9, 10-didehydro-N, N-diethyl-6-methyl- (8CI) (+) - L S D D—LYsergic acid diethylamide D—Lysergic acid Ν , Ν-diethylamide

Figure 5. NPROBE name search for LSD


HELLER AND M I L N E

Key

NIH/EPA

Chemical

Information

System

Structure

SCN 1

SCN 35

0 Ο

FG 219

o = p —ο —

Figure6.

Sample CIDS key codes


153

154

RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

OPTION? S P R O B E S P E C I F Y S T R U C T U R A L F E A T U R E C O D E A N D P E R M I S S I B L E M U L T I P L I C I T Y LIMITS N E X T S F C = SCN49 FOUND 428 C O M P O U N D S H A V I N G 1 OR M O R E O C C U R R E N C E S O F SCN49

N E X T S F C = SCN35 FOUND 277 C O M P O U N D

NEXT SFC = _ F I L E = 11, OPTION?

2 COMPOUNDS CONTAIN A L L

2 CODES

SSHOW 11

HOW M A N Y S T R U C T U R E S (E T O E X I T ) ? 1 T Y P E Ε T O T E R M I N A T E DISPLAY STRUCTURE 1 C A S R E G I S T R Y N U M B E R 6425-41-8 N I O S H R T E C S : QE06400,QE06700 C10H19NO C

C

C

C

Morpholine, 4 - c y c l o h e x y l Cyclohexylmorpholine N-Cyclohexylmorpholine 4-Cyclohexylmorpholine

Figure 7.

(8CI9CI)

CIDS key search for cyclohexyl morpholine compounds


10.

HELLER AND MILNE

NIH/EPA

Chemical

Information

System

155

M o l e c u l a r Weight (MW) and Formula (MF) Search In a d d i t i o n to searching f o r a p a r t i c u l a r f u n c t i o n a l group using the CIDS keys as shown above, i t i s p o s s i b l e to search f o r a compound, or a group of compounds, u s i n g molecular weight. The molecular weight search, shown i n F i g u r e 8, a l l o w s f o r e i t h e r a s p e c i f i c molecular weight, o r , as i s i n d i c a t e d i n the f i g u r e , a range o f molecular weights. In the p a r t i c u l a r example shown i n F i g u r e 8, the Merck Index i s being searched f o r a l l occurrences of compounds w i t h a molecular weight between 368 and 380. There are 167 such substances as can be seen i n the top p a r t of F i g u r e 8. T h i s i s too l a r g e a number and so i t was decided to t r y t o narrow or f i l t e r the search down to a smaller number u s i n g a molecular formula search. In t h i s case what was r e a l l y sought were a l l compounds which have two oxygen atoms and a molecular weight between 368 and formula (02) i s shown, o p e r a t i o n (INTERsect) between the f i l e of 167 compounds w i t h the c o r r e c t molecular weight range and the f i l e of 1484 having the c o r r e c t p a r t i a l formula. The r e s u l t of t h i s AND o p e r a t i o n i s a f i l e c o n t a i n i n g the 16 compounds i n the Merck Index which have a molecular weight between 368 and 380 as w e l l as e x a c t l y two oxygen atoms i n the molecule. A t the bottom of F i g u r e 8, the f i r s t of the 16 answers i s p r i n t e d o u t . T h i s compound, w i t h a molecular formula o f C21.H23.C1F.N.02 and a molecular weight of 375, i s H a l o p e r i d o l , which i s a drug used as a s e d a t i v e and t r a n q u i l i z e r . In the event t h a t there i s no i n t e r e s t i n c h l o r i n a t e d compounds, even though they may meet the molecular weight and molecular formula c r i t e r i a , a f u r t h e r molecular formula search may be conducted, as shown i n F i g u r e 9, f o r compounds w i t h 1-4 c h l o r i n e atoms. From F i g u r e 9, i t can be seen that there are 986 compounds w i t h 1-4 c h l o r i n e atoms i n the Merck Index f i l e . Since the requirement was f o r compounds t h a t d i d not c o n t a i n t h i s halogen atom, a Boolean NOT o p e r a t i o n between the 986 c h l o r i n e c o n t a i n i n g compounds and the 16 compounds p r e v i o u s l y found i s performed, as seen i n the center of F i g u r e 9. This r e s u l t s i n the removal o f three of the s i x t e e n substances, and o f the remaining t h i r t e e n , the f i r s t one, Androsta-3,5-dien-17-ol, 3-(cyclopentyloxy)-17-methyl-, (17.beta.), i s p r i n t e d out and shown a t the bottom of F i g u r e 9. T h i s , o f course, l i k e the other twelve i n the f i l e , does not c o n t a i n the c h l o r i n e that was present i n three of the answers to the f i r s t search shown i n F i g u r e 8. The a b i l i t y to i n t e r a c t and impose v a r i o u s l i m i t a t i o n s and f i l t e r s on searching i s a very powerful c a p a b i l i t y of the SANSS.


RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

OPTION? ^ TYPE MW OR RANGE, CR TO EXIT USER: 368-380 FILE = 4, 167 COMPOUNDS WITH MW 368-380 OPTION? MF CR TO EXIT, COMPLETE (C), PARTIAL (P), OR RANGED (R) MF? USER:£ THE NUMBER OF ATOM TYPES IS: J _ ENTER ATOM, FOLLOWED BY COUNT FOR EACH TYPE, E.G. C6. TYPE 1 IS: 02 FILE = 5, 1484 COMPOUNDS HAVING PARTIAL MF: 02

CR TO EXIT, COMPLETE (C), PARTIAL (P), OR RANGED (R) MF? USER: OPTION? INTER 4 5 FILE = 6 RESULTING SOURCE FILES WERE: 4 OPTION? SSHOW 6 HOW MANY STRUCTURES (E TO EXIT) ? J _ TYPE Ε TO TERMINATE DISPLAY STRUCTURE 1 C A S REGISTRY NUMBER 52-86-8 MERCK INDEX C21H23CIFN02 CL

C

c

c

C

C

c

F

C

C - Ο

C

C **C" C

1-Butanone, 4-(4-(4-chlorophenyl) -4-hydroxy-1-piperidinyl]-1-(4-fluoro phenyl)- (9CI)

Figure 8. Molecufor-weight range search


HELLER

ANDMILNE

NIH/EPA

Chemical

Information

System

OPTION? MF CR TO EXIT, COMPLETE (C), PARTIAL (P), OR RANGED (R) MF? USER: R THE NUMBER OF ATOM TYPES IS: 1 ENTER ATOM, FOLLOWED BY RANGE FOR EACH TYPE, E.G. C6,12. TYPE 1 IS: CL1 4 FILE = 7, 1&6 COMPOUNDS HAVING PARTIAL MF IN RANGE: CL1-4

OPTION? NOT 6 7 FILE = 8 RESULTING REFERENCE SOURCE FILES WERE: 6 OPTION? SSHOW 8 HOW MANY STRUCTURES (E TO EXIT) ? 5. TYPE Ε TO TERMINATE DISPLAY STRUCTURE 1 CAS REGISTRY NUMBER 67-81-2 MERCK INDEX

C25H3802

C ·· Ο

C

C ** C

0««C««C

C

C

C

C

C

C

C

Androsta-3,5-dien-17-ol, 3-(cyclopentyloxy)-17-methyl-, (17.beta.)- (9 CI)

Figure 9.

Sample of combination searches of MF, MW

with NOT logic


158

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

Nucleus - Ring Search (RPROBE) One of the f e a t u r e s of the CIS SANSS that has made the system u s e f u l i s the s t r u c t u r e of the f i l e w i t h respect to r i n g systems. The SANSS has a h i e r a r c h i c a l f i l e s t r u c t u r e that a l l o w s f o r r a p i d and inexpensive searching f o r s p e c i f i c r i n g s or r i n g systems. In F i g u r e 10, a l i s t of some of the commands used to generate s t r u c t u r e s are given. To show how the SANSS works and how one can use the v a r i o u s query modules, the remainder of the chapter w i l l be devoted to searching through the NIOSH TTECS data base f o r chemic a l s having an aromatic r i n g , s u b s t i t u t e d on ortho carbons w i t h c h l o r i n e and bromine r e s p e c t i v e l y . The f i r s t t h i n g that must be done i n order to perform such a search i s to b u i l d the 'query s t r u c t u r e that i s to be sought. This i s done w i t h the f i r s t few commands shown i n F i g u r e 11. The query s t r u c t u r e i n Figure 11 i s a c h l o r o bromo (ortho) probe search w i l l be conducte matic r i n g , s i n c e i t does not take i n t o account the nature of the s u b s t i t u e n t s . A l s o , s i n c e other s u b s t i t u e n t s on the benzene r i n g w i l l be p e r m i t t e d , i t i s necessary to r e s e t the s u b s t i t u e n t search l e v e l from EXACT (only two s u b s t i t u e n t s and these must be ortho) to 'IMBED (there must be two ortho s u b s t i t u e n t s at a minimum). The command to do t h i s i s EXIM, which i s short f o r EXact/lMbed s w i t c h . The search shown i n Figure 11 r e v e a l s that there are 2715 compounds i n the NIOSH RTECS f i l e that c o n t a i n at l e a s t t h i s r i n g p a t t e r n . To f i l t e r such p o t e n t i a l l y broad responses f u r t h e r , one can use CIDS keys searches and other such c o n s t r a i n t s as shown below. 1

1

1

1

Fragment Search (FPROBE) One f e a t u r e necessary to any s t r u c t u r e search system i s the a b i l i t y to search f o r atom-centered fragments. In a fragment search the user must s p e c i f y an atom and i t s neighbors. The exact (or g e n e r i c ) nature of the bonds between t h i s c e n t r a l atom and each of i t s neighbors i s then entered and a search i s conducted for a l l occurrences of such a fragment. I f a query s t r u c t u r e has already been generated, as was done i n F i g u r e 11, that s t r u c t u r e can be used by the SANSS program to generate and search f o r f r a g ments. There are u s u a l l y a number of atoms i n a query s t r u c t u r e that can be considered as c e n t r a l to a fragment. Hence, a request for a fragment probe of the s u b s t r u c t u r e shown i n F i g u r e 11 would l e a d to searches f o r s i x fragments, four of which would be the same ( i . e . atom centered fragments about atoms 3, 4, 5 and 6 are a l l the same, r e p r e s e n t i n g a carbon atom i n an aromatic r i n g attached to two other aromatic carbon atoms i n the r i n g and a hydrogen). Such fragments are not very s p e c i f i c , and so i t i s best to i d e n t i f y the atom centered fragment f o r which one wishes to search. In F i g u r e 12, atom number 1 i s s e l e c t e d and a search for a l l occurrences of a c h l o r i n e atom on an aromatic r i n g i s


HELLER

AND MILNE

NIH/EPA

Chemical

COMMAND

Information

System

EFFECT

A A T O M n1 m l

Insert an atom between atom n1 and atom m l .

A B O N D n1 m l

Insert a bond between n1 and m l .

A B R A N 11 at n1

A d d a branch of length 11 at atom n1.

A L I N K n1 11 m l

Insert a chain of length 11 between

A L T B D n1 m l

Define alternate bonds in the smallest

n1 and m l . ring containing n1 and m l as aromatic A R I N G n1 m l 11 and m l . CHAIN I

Create a chain of I atoms.

CLEAR

Erase the existing query structure.

C R I N G n1 11

Create a ring of 11 atoms including

DATOM η1

Delete atom η 1.

atom n1. D B O N D n1 m l

Delete the bond joining nl and m l .

MORGA

Renumber the query structure by the

NUC66

Create a structure of two fused

REG

Retrieve the structure corresponding

R EST

Negate the effect of the previous

RIΝ G I

Create a ring of I atoms.

S A T O M n1

Define the elemental nature of atom n1.

S B O N D n1 m l

Define the nature of the bond joining

S P I R O n1 11

Create a spiro-attached ring of

WISBD n1 m l

Define alternate bonds in the smallest

Morgan algorithm. six-membered rings. to a specific registry number. command.

n1 and m l . (11 +1) atoms at n1. ring containing η 1 and m l as double bonds.

Figure 10. Commands used to generate structures for searching


RETRIEVAL

ENTER NEW SELECTION

O F MEDICINAL

(H F O R HELP):

CHEMICAL

INFORMATION

32

C O L L E C T I O N S E L E C T E D : 32 OPTION? OPTION? RING OPTION? A B R A N 1 AT 1 1 AT 2 OPTION? S A T O M 7 SPECIFY E L E M E N T S Y M B O L =C L OPTION? S A T O M 8 SPECIFY E L E M E N T S Y M B O L = BR OPTION? A L T B D 1 2 OPTION? D 3..4

8BR2

5

1 . . 6 ? ? 7CL OPTION? EXIM SPECIFY S E A R C H L E V E L S T O B EC H A N G E D LEVELS = 4 OPTION? RPROBE C??C ? ? ? ? C

C?? ?

? ?

c??c

?

CONDITIONS O F S E A R C H CHARACTERISTICS T O B E MATCHED TYPE OF MATCH TYPE O F RING O R N U C L E U S EXACT NO HETEROATOMS EXACT SUBSTITUENTS A T 1 2 IMBED T H I S R I N G / N U C L E U S O C C U R S IN 2 7 1 5 COMPOUNDS FILE =

1,

2715

Figure 11.

COMPOUNDS C O N T A I N THIS

A ring-probe (RPROBE) benzene

RING/NUCLEUS

search for a disubstituted


10.

HELLER AND MILNE

NIH/EPA

Chemical

Information

System

161

performed. The r e s u l t of t h i s search i s a f i l e c o n t a i n i n g a l l 1618 compounds i n the NIOSH RTECS f i l e that c o n t a i n t h i s p a r t i c u l a r s t r u c t u r e fragment. A f t e r the fragment search i s conducted f o r the c h l o r o aromatic fragment, a s i m i l a r search i s performed on the fragment centered about atom 2, which contains a bromo s u b s t i t u e n t . This fragment probe (FPROBE) search, shown i n F i g u r e 13, r e s u l t s i n 229 occurrences of t h i s fragment i n compounds i n the NIOSH RTECS data base. Substructure Search (SUBSS) The Substructure Search o p t i o n i s an atom-by-atom, bond-bybond comparison between connection t a b l e s i n the data base and the connection t a b l e s corresponding to the query s t r u c t u r e . This time consuming, s e q u e n t i a l searc probe, fragment probe, are used as screens to speed up the process and reduce the c o s t . F o l l o w i n g the three separate searches done i n F i g u r e s 11-13, the next step i s to see which compounds i n the NIOSH RTECS data base c o n t a i n occurrences of a l l three. This i s done by a simple Boolean AND l o g i c combination of the three l i s t s of R e g i s t r y Numbers generated by the searches i n these F i g u r e s . The i n t e r s e c t i o n of the l i s t s , performed by the INTER command as shown i n F i g u r e 14, r e s u l t s i n 12 compounds meeting the c r i t e r i a of a l l three searches. However, not n e c e s s a r i l y a l l o f the 12 answers are p r e c i s e l y what i s wanted. This i s because the three searches i n F i g u r e s 11-13 are f o r " p i e c e s " of the s t r u c t u r e sought but the searches do not r e q u i r e these p i e c e s to be i n the same j u x t a p o s i t i o n as i n the query s t r u c t u r e . That i s , the three r e q u i r e ments comprise a necessary, but not s u f f i c i e n t c o n d i t i o n f o r an answer to the o r i g i n a l question. To secure an exact answer as to how many ( i f any) o f these 12 compounds meet the exact query s t r u c t u r e , i t i s necessary to perform a t r u e s u b s t r u c t u r e search (SUBSS) as i s shown i n F i g u r e 14. The r e s u l t o f the use of SUBSS shows t h a t only 7 o f 12 "answers" from the i n t e r s e c t i o n o f the three searches do have the bromine and c h l o r i n e ortho to one another on the benzene r i n g . Of the 7 answers, one i s shown i n F i g u r e 15. As i t turns out from i n s p e c t i o n of a l l 12 p r i o r answers (not shown h e r e ) , the other compounds r e t r i e v e d are meta s u b s t i t u t e d c h l o r o bromo aromatic compounds. Complete S t r u c t u r e Search (IDENT) The f i n a l SANSS module to be d e s c r i b e d i n t h i s chapter i s the search f o r a t o t a l or f u l l s t r u c t u r e , r a t h e r than a subs t r u c t u r e . T h i s module was designed p r i m a r i l y f o r the purpose of searching f o r and r e p o r t i n g s p e c i f i c chemicals as p a r t of the TSCA i n v e n t o r y r e p o r t i n g procedures. The f u l l s t r u c t u r e search, c a l l e d IDENT ( f o r IDENTity), has and w i l l continue to have


RETRIEVAL

OPTION?

O F MEDICINAL

CHEMICAL

INFORMATION

FPROBE 1

TYPE Ε T O EXIT F R O M A L L SEARCHES, Τ TO PROCEED T O NEXT FRAGMENT SEARCH FRAGMENT: 7CL????1C

6C

2C R E Q U I R E D O C C U R R E N C E S F O R HIT : 1 T H I S F R A G M E N T O C C U R S IN 1618 C O M P O U N D S F I L E = 2,

1618

COMPOUNDS C O N T A I N THIS F R A G M E N T

Figure 12. A fragment probe (FPROBE) atom attached

OPTION?

for a chlorine

FPROBE 2

TYPE Ε T O EXIT FROM A L L SEARCHES, Τ TO PROCEED TO NEXT FRAGMENT SEARCH FRAGMENT: 8BR????2C

1C

3C R E Q U I R E D O C C U R R E N C E S F O R HIT : 1 T H I S F R A G M E N T O C C U R S IN 229 C O M P O U N D S F I L E = 3,

229 C O M P O U N D S C O N T A I N T H I S F R A G M E N T

Figure 13. A fragment probe (FPROBE) for a bro mine atom attached to an aromatic carbon atom

OPTION?

INTER

FILE = 4 , SOURCE

1

2

FILES WERE:

OPTION?

3

RESULTING REFERENCES = SUBSSS

1

2

12

3

4

DOING SUB-STRUCTURE

SEARCH

T Y P E Ε T O EXIT FILE ITEM 10 S T R U C T U R E B E I N G S E A R C H E D HITS S O F A R 6 FILE = 5 ,

Figure 14.

SUCCESSFUL

SUB STRUCTURES

=

21609905

7

Intersection and substructure search of files de rived in Figures 11-13


HELLER

AND MILNE

STRUCTURE NIOSH R T E C S :

NIH/EPA

Chemical

7 CAS REGISTRY NUMBER TE70000

Information

System

4824-78-6

C10H12BrC12O3PS

CL * *

C * *

ο

C..C

* * C*****0**P++S * *

BR C #

C..C • * CL

ο * * C * C

Phosphorothioic acid, 0-(4-bromo-2, 5-dichlorophenyl) 0 , 0 — d i e t h y l ester (8CI9CI) Bromophos-ethyl Ethyl bromophos Filariol 60 Nexagan G

Figure 15.

One of seven substructure search hits


164

RETRIEVAL

O F MEDICINAL

CHEMICAL

INFORMATION

s p e c i f i c a p p l i c a t i o n t o TSCA a c t i v i t i e s . For example, a f t e r the f i n a l "grandfather i n v e n t o r y r e q u i r e d under s e c t i o n 8 of the Act i s p u b l i s h e d and made a v a i l a b l e , v i a the CIS, as w e l l as by other means, i t w i l l be necessary f o r p o t e n t i a l vendors o f a chemical t o determine i f the chemical they wish t o see o r manuf a c t u r e i s i n the Inventory and can thus be produced and marketed without e x t e n s i v e pre-manufacturing t e s t i n g . Use o f the IDENT search w i l l q u i c k l y r e v e a l i f the chemical i s i n the TSCA i n ventory. Of course, one can use the name search c a p a b i l i t i e s , but there i s no guarantee t h a t the name used by the manufacturer w i l l be i n the l i s t o f synonyms a s s o c i a t e d w i t h the i n v e n t o r y . The s t r u c t u r e shown i n F i g u r e 16 was generated u s i n g the standard SANSS s t r u c t u r e generation commands, such as those l i s t e d i n F i g u r e 10. The IDENT search was then invoked and a f t e r being t o l d that the s t r u c t u r e had the normal number of hydrogen atoms, cons i s t e n t w i t h normal v a l e n c e UDB. The s t r u c t u r e wa i d e n t i f i e r i n f o r m a t i o n , as w e l l as a number o f synonyms, one of which i s the TSCA C l e r i c a l Code Designation number f o r the substance. 11

SANSS-Data Base I n t e r f a c e s A s t r u c t u r e or a nomenclature search i s g e n e r a l l y only a means t o an end. The end i s o f t e n some data a s s o c i a t e d w i t h the s t r u c t u r e s found. In order t o f a c i l i t a t e r e t r i e v a l o f such i n f o r m a t i o n , an i n t e r f a c e between the CIS numeric data bases and the SANSS has been constructed. This a l l o w s f o r a search through the UDB f o l l o w e d by a data search (or r e t r i e v a l ) and permits one to answer such queries a s : * Do any ortho bromo-chloro aromatic compounds have a t o x i c i t y greater than 1.0 mg./kg? In the example shown i n Figure 17, the f i r s t three answers from the previous search are used to r e t r i e v e the t o x i c i t y data a s s o c i a t e d w i t h these compounds. The automatic i n t e r f a c e between the systems i s invoked by the command TSHOW and then the previous f i l e o f 7 CAS R e g i s t r y Numbers, generated by SUBSSS, are s p e c i f i e d , w i t h only the f i r s t three being p r i n t e d out upon request. Summary The NIH/EPA CIS has developed to the p o i n t where complex questions can be readilyanswered. The a b i l i t y to manipulate s t r u c t u r e and numeric data and e s t a b l i s h c o r r e l a t i o n s between the two should be o f c o n s i d e r a b l e value t o the EPA i n i t s work under the Toxic Substances C o n t r o l A c t , as w e l l as to s c i e n t i s t s i n g e n e r a l . The value o f the SANSS l i n k e d t o CNMR data has been r e c e n t l y shown (10), and no doubt other s t r u c t u r e - d a t a s t u d i e s


HELLER AND MILNE

OPTION?

NIH/EPA

Chemical

Information

System

ρ

10CL

70

?

+

?

+

8CL3??1??2P?50?11 ?

?

?

?

?

?

9 C L 4 0 60 ? ? 12 OPTION? TOTAL

IDENT

PROTON COUNT (P F O R P R O G R A M

TOTAL

F O R THIS S T R U C T U R E ESTIMATE)

PROTON COUNT BASED

:

IS

Ρ

UPON N O R M A L CONDITIONS

IS

8

ARE THERE A N Y ABNORMA WOULD

A F F E C T THI

PROTON COUNT FILE

FOR NOD

1 0 , T H I S S T R U C T U R E IS C O N T A I N E D

OPTION?

IN

1 COMPOUNDS.

S S H O W 10

STRUCTURE

1 CAS REGISTRY

TSCA CANDIDATE EPA PESTICIDES

-

EPA OHM/TADS:

CAMBRIDGE MERCK

LIST:

NUMBER

52-68-6

R001-5032

ACTIVE

INGREDIENTS:

57901

72T16519

XRAY

CRYSTAL:

52-68-6.01

INDEX

EPA PESTICIDES

-

EPA CHEMICAL

SPILLS

FDA/EPA

A N A L Y T I C A L R E F .STNDS.:

PESTICIDES

R E F .S T A N D A R D S :

PHS-149 C A R C I N O G E N S : NIOSH RTECS:

6780

48

C0147

TA07000 C4H8C1304P

CL

Ο

*

+

*

+

CL*C**C**P *0**C #

*

» CL

«

#

#

#

Ο

Ο

C Phosphonic acid,

(2, 2, 2-trichloro-1-hydroxyethyl)-,

dimethyl ester

(8C

I9CI) Agroforotox Anthon Bayer L

13/59

Chlorofos

Figure 16.

Example of IDENT

search for a complete molecule


166

RETRIEVAL

D A T A B A S E IS NOW

O F MEDICINAL

CHEMICAL

INFORMATION

RTECS

OPTION? R E T R I E V E N U M B E R I N G S Y S T E M ? CAS SOURCE? FILE 5 T H E R E WERE 7 D I S P L A Y HOW M A N Y ? (TYPE Ε T O EXIT) 3 CAS N U M B E R = 2104963 NIOSH N U M B E R = TE71750 O R L - R A T LD50: 1600 MG/K T F X : T X A P A 9 14,515,69 SKN-RBT LD50: 720 MG/K T F X : G U C H A Z 6.54,73 U N K - M A M LD50: 2000 MG/K T F X : 30ZDA9 -,335,71 Phosphorothioic acid, O — (4-bromo-2, 5-dichlorophenyl) O, O-dim ethyl ester (3CI9CI) C8H8BrC1203PS

CAS N U M B E R = 2720174 NIOSH N U M B E R = TB01850 O R L - R A T LD50: 35 MG/K TFX: A R S I M * 20,6,66 O R L - M U S LD50: 77 MG/K T F X : A R S I M * 20,6,66 Phosphonothioic acid, e t h y l - , 0-(4-bromo-2, 5-dichlorophenyl) O-ethyl ester (8CI9CI) C10H12BrC12O2PS

CAS NUMBER = 2720185 NIOSH N U M B E R = TB10700 O R L - R A T LD50: 73 MG/K T F X : A R S I M * 20,6,66 Phosphonothioic acid, methyl—, O—(4-bromo-2, 5-dichlorophenyl O-O-methylethyl) ester (9CI) C10H12BrC12O2PS Figure 17.

Example of NIOSH RTECS

toxicity data retrieval


10.

HELLER AND MILNE

NIH/EPA Chemical Information System

w i l l be undertaken now that the necessary groundwork has been laid. Acknowled gement s The authors wish to thank the f o l l o w i n g for t h e i r help and cooperation i n developing the CIS SANSS: R. J . Feldmann, W. G r e e n s t r e e t , M. Yaguda, M. Bracken, A . F e i n , G. Marquart, and J. Miller. Literature Cited 1. 2.

3. lished 4. 5. 6.

7.

8.

9. 10.

H e l l e r , S . R . , M i l n e , G.W.A., and Feldmann, R.J., Science, (1977), 195, 253. Feldmann, R.J., M i l n e G.W.A. Heller, S . R . F e i n Α . Miller, J . Α . , and (1977), 17, 157. The Interagency Regulatory L i a s o n Group (IRLG) was e s t a b 2 August, 1977 by the f o l l o w i n g four Agencies: EPA, FDA, OSHA and CPSC. EPA Order #2800.2, issued 27 May, 1975. Feldmann, R.J., and Heller, S.R., J. Chem. Doc., (1972), 12, 48. CIDS S t r u c t u r e Feature Key Code Manual is a v a i l a b l e from CIS P r o j e c t , Chemistry Department, Brookhaven N a t i o n a l Laboratory, Upton, Long I s l a n d , New York 11973. NIOSH, R e g i s t r y of Toxic E f f e c t s of Chemical Substances (RTECS), 1977. A v a i l a b l e from the US Government P r i n t i n g O f f i c e , GPO Order Number 017-033-0027101; $17.50 per copy USA: $21.88 per copy non-USA. Bracken, Μ., D o r i g a n , J., Hushon, J., and Overbey, II, J., MITRE Reprt MIR-7558 to CEQ, June 1977. Two volumes en titled "Chemical Substances Information Network (CSIN)". NLM Fact Sheet for the T o x i c o l o g y Information Program, January 1978. M i l n e , G . W . A . , Zupan, J., Heller, S . R . , and Miller, J.A., A n a l . Chim. A c t a , In press (1978).



167

11

An

Integrated System for C o n d u c t i n g

Biological

Chemical

and

Searches

T. M . DYOTT, A. M . EDLING, C. R. GARTON, W. O. JOHNSON, P. J. McNULTY, and G. S. ZANDER Rohm and Haas Company, Norristown Road, Spring House,PA19477 Over the past seve t Roh d Haa Compan hav been developing a computerize tion system called ACCI ( A g r i c u l t u r a Computerize Information System)(1). In this paper we will d e s c r i b e the chemical and biological search capabilities which we have built i n t o ACCIS. ACCIS Design

Criteria

ACCIS was developed in order t o : 1.

Accomodate the growing amount of data which r e s u l t e d from expanding biological s c r e e n i n g programs.

2.

Facilitate communication of screening r e s u l t s to r e s e a r c h e r s , a d m i n i s t r a t o r s , and o u t s i d e collaborators.

3.

Reduce the time our biologists spent t r a n s c r i b ing, e x t r a c t i n g , and r e p o r t i n g screening results.

4.

Enhance the value of the s t o r e d screening r e s u l t s by making them readily a v a i l a b l e .

To meet these o b j e c t i v e s we decided that the system must: 1.

c o n t a i n not only the biological screening r e s u l t s , but a l s o the chemical s t r u c t u r e s , reference d a t a , and p e r t i n e n t chemical d a t a , e.g., solubility and purity information.

2.

produce a v a r i e t y of current awareness r e p o r t s on standard 8 1/2 X 11 paper, or 3 X 5 or 5 X 8 c a r d s , and that those r e p o r t s should c o n t a i n h i g h q u a l i t y structural diagrams whenever a p p r o p r i a t e . 0-8412-0465-9/78/47-084-168$05.00 © 1978 American Chemical Society In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978.

11.

DYOTT ET

3.

AL.

Chemical

and Biological

Searches

169

provide a convenient mechanism f o r conducting a wide v a r i e t y of chemical and/or b i o l o g i c a l searches.

System O r g a n i z a t i o n ACCIS i s thoroughly i n t e g r a t e d i n t o a everyday o p e r a t i o n of our screening programs. The flow of i n f o r m a t i o n i n t o ACCIS i s diagrammed i n Figure 1. When our chemists synthesize a compound they complete a compound s u b m i t t a l form, g i v i n g the e m p i r i c a l formula, s t r u c t u r a l diagram, chemical name, chemist's name, notebook r e f e r e n c e , department, date, v a r i o u s p h y s i c a l p r o p e r t i e s , screening p r i o r i t i e s , and any s p e c i a l i n s t r u c t i o n s . The chemist then takes the s u b m i t t a l form and the sample i t s e l f to the Screening Information Center. There the i n f o r m a t i o n i s reviewed and entered i n t o the system v i a a chemical t y p e w r i t e r (a modified IBM MCST) sent to the appropriat produced t r a n s m i t t a l sheet which provides the b i o l o g i s t s w i t h the s t r u c t u r a l diagram, u s e f u l p h y s i c a l property i n f o r m a t i o n , and any s p e c i a l i n s t r u c t i o n s . The b i o l o g i s t s then screen the compound, r e c o r d i n g t h e i r f i n d i n g s on 2-part carbonless forms. They keep the f i r s t copy as a l e g a l r e c o r d , w h i l e the second copy i s returned to the i n f o r m a t i o n center where the data are keypunched and read i n t o the system. Whenever data are entered, v a r i o u s current awareness r e p o r t s are a u t o m a t i c a l l y generated which keep the chemists, b i o l o g i s t s , and t h e i r management i n formed and a l l o w them to maintain hardcopy f i l e s . A t y p i c a l ACCIS r e p o r t , the h e r b i c i d e current awareness r e p o r t , i s shown i n Figure 2. (The organism names have been replaced by the l e t t e r s B-L f o r c o n f i d e n t i a l i t y reasons.) AM and AD are average c o n t r o l data f o r a l l monocot and a l l d i c o t s p e c i e s , r e s p e c t i v e ly. The number of screening programs f l u c t u a t e s as new programs are i n i t i a t e d and o l d ones are terminated, but i s g e n e r a l l y i n the range of 8-12. Each screen may i n t u r n i n c l u d e anywhere from 1 to 15 d i f f e r e n t organisms, t r e a t e d under v a r i o u s c o n d i t i o n s and dosages. This v a r i a b i l i t y makes i t e s s e n t i a l that the b i o l o g i s t s i n each area work c l o s e l y w i t h the i n f o r m a t i o n s p e c i a l i s t to design both t h e i r data c o l l e c t i o n forms and the v a r i o u s r e p o r t s they r e q u i r e . Our emphasis i s on meeting the researcher's needs r a t h e r than s i m p l i f y i n g the programming. As a r e s u l t ACCIS: 1.

i s a h i g h l y customized

system.

2.

c o n s i s t s of w e l l over 100 programs, t o t a l i n g approximately 250,000 l i n e s of code.

3.

enjoys extremely

strong user

support.


170

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

BIOLOGISTS

BIOLOGY DATA

BOUND PAGE

I

FORM

Figure 1. Flow of information into ACCIS


FOR LEGAL


1 3

3

1650 N 0S

3

2

2

RATE ( #/A) 8 4 4

RATE ( */A> 8 4 4

7

4

TEST DATE 09/18/74 09/18/74 09/18/74

I O

TYPE TEST PRE PRE PCST

TYPE TEST PRE PRE PCST

TYPE RATE ( */A ) TEST PRE 8 PRE 4 PCST 4

RH 1652 C H CIN 0 S MAYBRIDGE

TEST DATE 09/18/74 09/18/74 09/18/74

s

RH 1651 C|oH F N4 MAYBRIDGE

TEST DATE 09/18/74 09/18/74 09/18/74

MAYBRIDGE

RH C*H

*********•*•*•**•••*•• •COMPANY CONFIDENTIAL* **********************

0

Β 0

D 0

\

F - * - F

F I

/

•—NH—Ν=·—·=Ν

• Ν S Ο · Il II il II J • ·—NH—·—NH—·—·—· \ / I J Κ AM G 0 0 0

ROHM ANO HAAS COMPANY CURRENT AHA«ENESS REPORT HERBICIDE

STA TUS D D 100 D

AD

0

Β 0

0

C

ο ο

ε

Figure 2.

-

ο

0

G

/

A typical ACCIS

AM

\

h

Ο

\

/

J

report format

C

C l

0

Κ 0 0

L 99 -

STA TUS D D D

STA TUS J Κ L F I Ε F AM G C D AD Β 0 40 40 60 7C 0 90 0 90 44 90 100 99 0 30 60 40 30 95 100 32

AD

AREA

03/23/78

172

RETRIEVAL

OF

MEDICINAL

CHEMICAL

INFORMATION

The chemical and b i o l o g i c a l i n f o r m a t i o n i n ACCIS i s s t o r e d i n a number of computer f i l e s . The b i o l o g i c a l , m i s c e l laneous chemical, and reference i n f o r m a t i o n i s s t o r e d i n an IMS data base. The s t r u c t u r a l diagram, as entered on the chemical t y p e w r i t e r , and the chemical name are stored i n standard v a r i able record l e n g t h f i l e s . In order to s t o r e the chemical s t r u c tures i n a machine i n t e l l i g i b l e , and t h e r e f o r e searchable, manner we incorporated the Chemical A b s t r a c t s Servic e (CAS) R e g i s t r y I I system i n t o ACCIS. The s t r u c t u r e s are s t o r e d i n a connection t a b l e f i l e and a fragment f i l e i s generated which improves the e f f i c i e n c y of the substructure search system. In a d d i t i o n there are a number of a u x i l i a r y f i l e s which d e s c r i b e the b i o l o g i c a l screens and are used to v a l i d a t e the b i o l o g i c a l data, a l l o w a b b r e v i a t i o n s i n the data base to be expanded i n r e p o r t s (data d i c t i o n a r i e s ) , and supply d i s t r i b u t i o n l i s t s f o r v a r i o u s r e p o r t s . The t o t a s t e a d i l y s i n c e ACCIS m i l l i o n characters. 1

Search C a p a b i l i t i e s We found that i n a d d i t i o n to current awareness r e p o r t s we needed to be able to produce r e p o r t s based on v a r i o u s c r i t e r i a , e.g., s u b s t r u c t u r e , b i o l o g i c a l a c t i v i t y , t e s t date, and/or source. T y p i c a l questions might be: 1.

What 5-halo i s o t h i a z a l o n e s have we made?

2.

What compounds have we screened which c o n t r o l >80% of weed XYZ when a p p l i e d at 2 l b s / a c r e preemergence?

3.

What are the f u n g i c i d e screening r e s u l t s f o r the compounds we obtained from KLM corporation?

4.

What 4 - n i t r o diphenyl-ethers have we made which c o n t r o l >80% of weed RST when a p p l i e d at 4 l b s / a c r e postemergence?

5.

What compounds were screened f o r i n s e c t i c i d a l a c t i v i t y during December 1977?

D i f f e r e n t types of r e p o r t s are a l s o c a l l e d f o r . We might need j u s t the s t r u c t u r e s and reference i n f o r m a t i o n , or s t r u c tures and the screening r e s u l t s from a p a r t i c u l a r area, or s t r u c t u r e s and the screening r e s u l t s from s e v e r a l areas. Since a l l of our common questions are compound o r i e n t e d we designed a modular search system as shown i n Figure 3.


In Retrieval of Medicinal Chemical Information; Howe, W., et al.; ACS Symposium Series; American Chemical Society: Washington, DC, 1978. Figure 3.

BIOLOGICAL SEARCH PROGRAM

CHEMICAL SEARCH PROGRAM

STRUCTURE AND ALL BIOLOGICAL AREAS REPORT PROGRAM

STRUCTURE AND SPECIFIC BIOLOGICALl AREA REPORT PROGRAMS

Flow diagram of modular ACCIS search system

SELECTED OMPDS.

ρ

STRUCTURE REPORT PROGRAM

RETRIEVAL

174

O F MEDICINAL

CHEMICAL

INFORMATION

A s u i t a b l e chemical search program f o r CAS R e g i s t r y I I f i l e s had already been developed by CAS, w h i l e the v a r i o u s r e p o r t programs are modified v e r s i o n s of current awareness r e p o r t programs we have p r e v i o u s l y developed. The only major new program we needed was one f o r searching the b i o l o g i c a l and r e f e r e n c e i n f o r m a t i o n contained i n the IMS data base. Biological

Search

The b i o l o g i c a l data we need t o search i s contained i n an IMS data base, which has a h i e r a r c h i c a l s t r u c t u r e , as shown i n Figure 4. This h i e r a r c h i c a l s t r u c t u r e allows you t o have any number of t e s t areas w i t h i n a compound, any number of t e s t dates w i t h i n a t e s t area, any number of t e s t types w i t h i n a t e s t date, etc. (There i s of course more d e t a i l e d i n f o r m a t i o n w i t h i n each segment of the data bas We developed a searc search c a p a b i l i t y . I t allows us to q u a l i f y the search or any p i e c e (or pieces) of i n f o r m a t i o n i n the data base and has cons i d e r a b l e Boolean l o g i c c a p a b i l i t i e s . For example, i f we were i n t e r e s t e d i n compounds w i t h i n the range RH-60000 to RH-80000 which were a c t i v e a g a i n s t fungus ABC o r DEF, but d i d not i n j u r e crop XYZ a t a r a t e of 4 l b s / a c r e , we would encode the question as :

(RH>60000*RH F l Registry where structure > F2 facility 118 where structure = SI components 119 elements function 118 file organization Number, CAS 147 full-structure (FSS) of Toxic Effects of Chemical inventory subsystem Substances (RTECS) 63 keys, development of file assessment 71 on-line substructure Relational data base management on-line system system (RDBMS) 123 operation programs for particular Report generation 136 parameters Report generator 197 products, typical results file Research proposals, analysis of 198


184 183

202 198 169 201 203 198 174 197 172 197 122 102 122 122 121 123 211 212 197 209 194 193 91 98 19 121

230

RETRIEVAL

Search ( es ) ( continued ) strategies 89,196 substructure 198 of CA 217 subsystem 123 system biological 174 fragment-coded 108 modular 173 three-phase 124 on Topology (SRT) 217 types 90,91 Searching chemical structure 107,108 chemical substructure 107, 108 the file 211 literature 217 procedures 193 substructure (SS) 124 Secondary data 8 software 25 sort parameter 46 Security, system 137 Software DR & D CIS 215 function of 125 primary 24 secondary 25 Special atoms, relationships among 187 atoms, types of 187 structure feature identification (SSFID) processor 217 Specificity 50 SRT (Search on Topology) 217 SS (see substructure search) SSFID (special structure feature identification) processor 217 Stand-alone computer system 109 Structural diagrams for retrieved substances 69 Structure data base 108 -drawing operation 113 entry system 108, 109 generated from SANSS structure generation commands 165 input, graphical Ill keys, Edgewood CIDS 147 and nomenclature search system (SANSS) 144,145,147 data base interfaces 164 structure generation commands, structure generated from 165 representation, internal Ill search (IDENT), complete 161 table 126

O F MEDICINAL

CHEMICAL

INFORMATION

SUBSS (see substructure search system ) Substance data base 136 Substructure data 123 retrieval capabilities 97, 98 search (es) (SS) 124,198,212 of CA 217 C H E M LI NE 65 approach 66 -RTECS 68, 71 minicomputer 126 storage format 127 on-line 194 query for nucleosides, example of prestored component pieces 213 subsystem 123 System (SUBSS) 161, 176 CAS 176 tion of table System configuration design criteria operation organization, ACCIS security support

125 126 24 23 25, 26 169 137 24

Τ

Term index to protocol file 87 Test-data transmission from producer to computer master file, automa tion of 20 Testing programs, cost-effective 90 Text data bases 53 Token distribution functions 46 Token frequency distribution func tions, nature of 48 TOXicology information on-LINE (TOXLINE) 62 on-line retrieval of 62 -TOXBACK 62 TOXLINE (see TOXicology infor mation on-LINE) Three-phase search 124 TRAC 208 Transaction file 97,100,133 data base 100 record format 101 Transaction inventory 100 (TSCA) Inventory Candidate List, Environmental Protection Agency's Toxic Substance Control Act 58 Type distribution functions 46


231

INDEX

User acceptance 24 υ User language 121 U-number Ill UDB (Unified data base), CIS 144,146 Unified data base (UDB), CIS 144,146 V Unspecified bond 187 Varian V74 137,140 Upjohn Company, on-line chemical and biological information system at 107 W Use(s) Walter Reed Army Institute of combinations 42 Research (WRAIR) 181 per compound, frequency sort for .. 47 Chemical Information Retrieval -compound parameter 42 System (CIRS) 181 correlations 54 subsystems 181 data file, medical 40 distribution file 46, 47 Warner-Lambert/Parke-Davis (WL/PD) 132 frequencies 53 CAS files at 136 -oriented data 44 CAS Registry III Integrated pairs 44 dominant 4 having significant specificity parameter 52 WL/PD (see Warner-Lambert/ Parke-Davis ) token 40 46 occurrences 53 Word distribution table 205 qualifier subsets of 55 Work stations NSC assignment processor 208 type(s) 40,42 classes of 46 WRAIR (see Walter Reed Army Institute of Research) descriptors 53


Retrieval of Medicinal Chemical Information

Information Retrieval

Information Retrieval

The geometry of information retrieval

Computational Information Retrieval

Visualization for Information Retrieval (The Information Retrieval Series)

Readings in Information Retrieval

Information Retrieval Interaction

Introduction to Information Retrieval

Modern Information Retrieval

Human Information Retrieval

Modern information retrieval

Information Retrieval Technology

Modern Information Retrieval

Introduction to Information Retrieval

Information Retrieval 001.ps.gz

Visualization for Information Retrieval (The Information Retrieval Series)

New Directions in Cognitive Information Retrieval (The Information Retrieval Series)

Information Retrieval Architecture and Algorithms

Information Retrieval - SciFinder, 2nd edition

An Introduction To Information Retrieval

Advanced Topics in Information Retrieval

Introduction to Information Retrieval (Draft)

Information Retrieval und Künstliche Intelligenz

Multidisciplinary Information Retrieval - IRFC 2011

The Modern Algebra of Information Retrieval

The Modern Algebra of Information Retrieval

The modern algebra of information retrieval

Intelligent Document Retrieval : Exploiting Markup Structure (The Information Retrieval Series)

Music Retrieval (Foundations and Trends in Information Retrieval)

Music Retrieval (Foundations and Trends in Information Retrieval)

Retrieval of Medicinal Chemical Information

Information Retrieval

Information Retrieval

The geometry of information retrieval

Computational Information Retrieval

Visualization for Information Retrieval (The Information Retrieval Series)

Readings in Information Retrieval

Information Retrieval Interaction

Introduction to Information Retrieval

Modern Information Retrieval

Human Information Retrieval

Modern information retrieval

Information Retrieval Technology

Modern Information Retrieval

Introduction to Information Retrieval

Information Retrieval 001.ps.gz

Visualization for Information Retrieval (The Information Retrieval Series)

New Directions in Cognitive Information Retrieval (The Information Retrieval Series)

Information Retrieval Architecture and Algorithms

Information Retrieval - SciFinder, 2nd edition

An Introduction To Information Retrieval

Advanced Topics in Information Retrieval

Introduction to Information Retrieval (Draft)

Information Retrieval und Künstliche Intelligenz

Multidisciplinary Information Retrieval - IRFC 2011

The Modern Algebra of Information Retrieval

The Modern Algebra of Information Retrieval

The modern algebra of information retrieval

Intelligent Document Retrieval : Exploiting Markup Structure (The Information Retrieval Series)

Music Retrieval (Foundations and Trends in Information Retrieval)

Music Retrieval (Foundations and Trends in Information Retrieval)

Recommend Documents