220th National Meeting
Washington, DC, August 20-24, 2000

A. H. Berks, Program Chair

NOTE: Presentations with slides or other materials available are marked with . A few files are in Word or HTML format. The type of file available will be indicated for each presentation (PowerPoint = ppt, Adobe/PDF = pdf, Word = doc, HTML = htm).

SUNDAY MORNING

SOCED/CINF Chemical Information Symposium

Section A

Virtual High-Throughput Screening
Receptor-based approaches - I

Grand Hyatt -- Franklin Square
Cosponsored with Division of Computers in Chemistry, and Division of Medicinal Chemistry
O. F. Güner, Organizer
M. Waldman, Presiding

8:25 — Introductory Remarks.

8:30 — 1. Progress toward a protein-ligand scoring function for fast docking. Marvin Waldman, Paul Kirchoff, Jeff Jiang, and C.M. Venkatachalam, Molecular Simulations Inc, 9685 Scranton Road, San Diego, CA 92121, marvin@msi.com

A novel scoring function for protein-ligand interactions developed to predict the binding affinity of drug candidate compounds docked into an active site will be presented. The function has been parameterized using a wide variety of protein-ligand systems and can predict binding affinities for systems for which other scoring functions perform poorly. Although it uses far fewer parameters than previously developed scoring functions, equivalent accuracy has been achieved with improved parameter transferability. The function is rapid to compute while accounting for steric or van der Waals interactions, complimentary electrostatic interactions, and desolvation effects. Comparison with prior literature scoring functions will be presented along with examples illustrating applications to compound prioritization and selection.

9:00 — 2. Computational geometry analysis of protein-ligand complexes. Alexander Tropsha, and Jun Feng, School of Pharmacy, University of North Carolina at Chapel Hill, CB # 7360, Beard Hall, School of Pharmacy, Chapel Hill, NC 27599-7360, Fax: 919-966-0204, alex_tropsha@unc.edu

Delaunay tessellation (DT) has been applied to a diverse set of x-ray characterized high resolution structures of ligand receptor complexes. Using nonhydrogen atom coordinates, DT generates an aggregate of space filling irregular tetrahedra with atoms as vertices. Tetrahedra formed naturally at the ligand-receptor interace incorporate both ligand and receptor atoms. Four body statistical potentials and a scoring function for ligand receptor recognition has been developed using various nuber of atom types for bot ligands and receptor proteins. Preliminary studies indicate that this novel scoring function affords a high correlation between experimental and calculated binding constants: the correlation coefficient is 0.7 for 57 diverse complexes, and 0.85 for 16 serine protease-inhibitor complexes. This scoring function can be successfully used to rank compounds resulting from high-throughput docking studies.

9:30 — 3. Putting the horse before the cart: Analysis and optimization of structure-based virtual screening protocols. Andrew C. Good, Daniel L. Cheney, William E. Harte, Yi Li, Stanley R. Krystek, Donna A. Bassolino, John S. Tokarski, Terry R. Stouch, Yaxiong Sun, Malcolm E Davis, Deborah Loughney, Jonathan S. Mason, and Doree F. Sitkoff, Structural Biology and Modeling, Bristol-Myers Squibb, 5 Research Parkway, Wallingford, CT 06492, Fax: 203-677-7702, gooda@bms.com

Slides (ppt)
There has been much research into the development of new scoring functions for structure-based virtual screening. While some advances have been made in improving virtual screening results through this approach, in general progress has been limited. Here we highlight results obtained from DOCK studies designed to improve virtual screening via analysis of the screening phases that occur before scoring. Conformational analysis studies were undertaken on diverse PDB ligands using a variety of techniques. Search methods were compared via their ability to reproduce conformers close to the bioactive structure at sampling levels typically employed in virtual screening. In addition, 5 different target proteins each with associated active compound data sets were used to analyze the effect of docking variables such as ligand flexibility, site point definition and node sampling levels. The ranking of these active compounds when combined with a set of ~10000 "noise" compounds was used to compare screening enrichment levels and hence better determine optimum DOCK search paradigms. The results of these studies are discussed and their implications for the direction of future virtual screening research are highlighted.

10:00 — 4. Effect of protein model accuracy on virtual high-throughput screening experiments. Mariusz Milik, Lisa Yan, Krzysztof Olszewski, Azat Badretdinov, and Sándor Szalma, Life Sciences, Molecular Simulations Inc, 9685 Scranton Rd., San Diego, CA 92121-3752, Fax: 858-458-0136, mmilik@msi.com

The amount of protein structures determined can not keep pace with the amount of possible targets being generated during the genome sequencing efforts. We have recently introduced and validated a high-throughput modeling pipeline for generating protein structures based on genomic sequences. We have shown that the structures generated with this pipeline can be efficiently used for deriving annotations undetectable with more conventional bioinformatics tools.

We have constructed a library of cross-modeled protein structures chosen from diverse families and have performed docking experiments of native ligands to a set of model structures. We examined how reliably the protein structures generated from GeneAtlasTM pipeline can be used in docking and screening experiments, i.e.: what the quality of the predicted docked structures is compared to the X-ray crystallographic structures and how accurately the binding constants can be predicted by a recently developed scoring function.

The prospects for reverse virtual high-throughput screening and the extension of VHTS protocols for model annotation is also discussed.

10:30 — 5. Structure based design of inhibitors for matriptase. Istvan J. Enyedy⁽¹⁾, Sheau-Ling Lee⁽²⁾, Chen-Yong Lin(²⁾, Robert B. Dickson⁽²⁾, and Shaomeng Wang⁽³⁾. (1) Drug Discovery Program, Georgetown University Medical Center, 3900 Reservoir Road, N.W, Washington, DC 20007-2197, istvan@giccs.georgetown.edu, (2) Lombardi Cancer Center, Georgetown University Medical Center, (3) Lombardi Cancer Center and Department of Oncology, The Structural Biology and Cancer Drug Discovery Program

Matriptase, a novel membrane-integrated trypsin-like serine protease, may be involved in tissue remodeling and metastasis of breast cancer. Potent and selective matriptase inhibitors have the potential to be used in elucidating the role of matriptase in breast cancer invasion. The structure of matriptase obtained from homology modeling was used for structure-based database search of the NCI database of drug-like molecules in order to select potential inhibitors. We identified several compounds that are potent inhibitors of matriptase and have good selectivity when compared to trypsin, uPA and thrombin. The most potent compound has Ki value 191 nM in inhibition of matriptase and is also the most selective inhibitor among the all inhibitors discovered. Flexible ligand-flexible receptor docking in explicit water was used to obtain the binding mode of the most potent inhibitors to matriptase. Our data suggest that homology modeling combined with 3D-database search is a powerful tool in identifying potent inhibitors even when the X-ray structure of the target protein is not known.

11:00 — 6. Virtual high throughput screening: Coarse-grain parallelization applied to ligand-receptor docking of large datasets. Joe Leonard, and Malcolm Cline, Development, Tripos, Inc, 1699 South Hanley, St. Louis, MO 63144, Fax: 314-647-9241, jleonard@tripos.com, mac@tripos.com

Rational drug design based on crystallographic receptor structures is a well known technique in computational chemistry that has led to the development of automatic docking software such as DOCK, FlexX, and Gold. At the same time there has been a paradigm shift in the pharmaceutical industry from one-at-a-time to robotic and combinatorial synthesis of large compound libraries for screening. This shift is putting pressure on docking programs to become fast and accurate enough to computationally evaluate binding strengths or docking feasibility of large libraries of compounds so that library designs may be improved prior to synthesis. One solution, addressed here, is to use existing software with large numbers of processors, since the docking of databases is readily parallelizable.

Section B

Copyright In the Digital Environment: Current Issues and Future Changes

Grand Hyatt -- Constitution Ballroom A
Cosponsored with Division of Chemistry and the Law
B. Lawlor, Organizer
D. Lide, Presiding

8:55 — Introductory Remarks.

9:00 — 7. Introduction to copyright law for the chemical information professional. Peter A. Cicala, Lerner, David, Littenberg, Krumholz & Mentlik, LLP, 600 South Avenue West, Westfield, NJ 07090, Fax: 908-654-7866, pcicala@ldlkm.com

In the computer age, it is imperative for the chemical information professional to understand the fundamental principles and framework of copyright laws. In order to fully appreciate the impact such laws have on chemical information technology, it is important to be familiar with the scope of copyrightable subject matter, the difference between patent, trademark, and copyright protection, fair use of otherwise copyrighted works, the licensing of copyrighted works, works for hire, when copyrights vest, and enforcing copyrights. These fundamental copyright principles will be used to govern how information is gathered, used, and commercialized in the computer age.

9:30 — 8. Database protection in the United States and abroad. Dan Duncan, Information Policy Consultant, 114 East Maple Street, Alexandria, VA 22301, dcd1952@bellatlantic.net

Since 1996, the world at large has continued to move toward developing new protections for databases outside copyright law. With the European Union leading the way, enactment or consideration of a new database protection law has been at issue at the World Intellectual Property Organization and in countries as diverse as the former Eastern Bloc nations, Mexico and Canada. Yet in the United States, the debate over how to protect databases has grown ever more rancorous - fueled in large part by the enormous economic stakes involved - and seems unlikely to be resolved in the near future. The presentation will review the world legal model that has begun to emerge and compare it with the various proposals considered thus far in the United States. Finally, there will be an exploration of the consequences for both industry and users should Congress fail to establish a new federal protection statue for databases.

10:00 — 9. Tap dancing on the quick sand: Cautionary copyright tales from academe. F. Bartow Culp, Libraries, Purdue University, Mellon Library of Chemistry, 1538 Wetherill, West Lafayette, IN 47907-1538, Fax: 765-494-1579, bculp@purdue.edu

TICE (The Internet Changes Everything) - especially in the potent combination of easily available electronic information resources, e-mail, and the "virtual classroom". The power to capture, alter, and disseminate information with a few mouse clicks is straining the traditional envelope of copyright protection, a situation that frequently leads members of the academic community to venture into unfamiliar and legally dangerous territory. The unauthorized inclusion of publications in a web site and the wholesale scanning of documents for course reserves are examples of possible copyright law violations. Also, the increasing availability of electronic journal articles has sharpened the debate over who actually owns them - the publisher, the author, or even the author's sponsoring institution. The purpose of this talk is to present some key copyright issues facing academicians, and to illustrate these issues with practical examples and solutions.

10:30 — 10. Why Copyright still matters in the Web Millennium. Robert D. Bovenschulte, Publications Division, American Chemical Society, 1155 16th Street, N.W, Washington, DC 20036, Fax: 202-872-6060, rxb96@acs.org, and Robert J. Massie, Chemical Abstracts Service, P.O. Box 3012, Columbus, OH 43210, Fax: 614-447-3765, rmassie@cas.org

Copyright has the reputation of being a dull subject. Whether that reputation is historically deserved or not for printed publications, it is definitely wrong in the Web era. Copyright has become an important subject gaining increasing attention. And, while the WEB has made powerful technologies available to anyone who wishes to bend or break the copyright law, the law and the related responsibilities remain fundamentally unchanged. Scientific journals have surged onto the Web in recent years. They offer faster publishing, easier access, and greater functionality than their printed precursors. But online journals raise questions about publishers' policies and practices, and many of them relate to copyright. Additionally, with regard to databases the Supreme Court made it clear in the 1991 Feist ruling that the "sweat of the brow" doctrine is not valid for copyright protection - copyright is intended to protect intellectual property resulting form creative work. As a result, opinions about copyright for online information are widely divergent. Not suprisingly, viewpoints differ from one constituency to another, since the interests of authors, researchers, librarians, and publishers are sometimes at odds. Moreover, changes in copyright, or in publishers' policies with respect to it, are often identified as a way to accomplish other goals - like lower prices - which are not strictly related to copyright. Issues of control and usage abound in the digital revolution that is transforming scientific communication. And there are, understandably, so many worries and uncertainties about the future implications of any changes that may be made today. This presentation will examine these issues entailing a broad consideration of copyright for Web Journals and SciTech databases. The focus will be on the salient questions and challenges that publishers are confronting from their customers. The ACS position on copyright with regard to online journals, and the creative intellectual effort by the CAS scientists who build the CAS databases will be explained. Speculations will be offered about future policies that might be considered by ACS and other scientific publishers, together with common-sense approaches to customer agreements that have worked to preserve copyright while giving information users the accessibility essential to research.

11:00 — 11. Recent developments in Copyright Law regarding the use of electronic information. Rachel Goslins, Office of Policy and International Affairs, Copyright Office, Library of Congress, 101 Independence Avenue, S.E, Washington, DC 20540, Fax: 202-707-8366, rgos@loc.gov

With the enactment of the Digital Millennium Copyright Act of 1998(DMCA), several new and important tenets in copyright law were established. These changes affect how electronic information will be accessed, used and licensed by the library, academic and research communities in the upcoming future. This presentation will provide a brief overview of the significant provisions in the DMCA, and focus in particular on two recent developments - a study completed by the Copyright Office recommending an amended exemption for distance education uses over digital networks, and a rule-making by the Office currently underway to determine possible exemptions to the anticircumvention provisions of the DMCA.

11:30 — 12. Extenuating circumstances: Intellectual Property and the politics of information. Dick Kaser, National Federation of Abstracting and Information Services, 1518 Walnut Street, Suite 307, Philadelphia, PA 19102, Fax: 215-893-1564, kaser@nfais.org

Information has never been without its political context, but the politics have rarely been so thick as they are now. "Information wants to be free" - free of structure, free of unnecessary constraints, even free of charge. Facts want to remain unprotected. Fair use wants to continue. Government information wants to remain in the public domain. The exclusive rights of the copyright holder want to remain in force. Libraries want to remain resource centers and archives. Publishers want to remain producers and distributors. Government agencies want to remain active disseminators....One, perhaps inevitable, result of our transition to a networked digital information world is political conflict among the stakeholders. But is rhetoric replacing reason? Does motherhood and apple pie actually make for the best scientific information policy? In our society's apparent zeal to democratize information services, we run the real risk of destroying the very balance between public interest and private ownership that has seen science through the 20th century. Shall we drum out of existence the "tyrannical" commercial publishers and "high-priced" value-added information service providers? Shall we let the general public pick up the full tab of reinventing and maintaining digital information services that actually work? These and other issues will be discussed in this presentation.

SUNDAY AFTERNOON

Section A

1:25 — Introductory Remarks.

1:30 — 13. Investigation of in silico vs. in vitro high throughput screening for hit identification. West Sussex RH12 5AB, United Kingdom, Fax: +44 (0)1403 323307, lewis.whitehead@pharma.novartis.com

Slides (pdf)
In vitro high throughput screening (HTS) methodologies have had a radical impact on the hit identification process in the pharmaceutical industry. However, there are many caveats to the current reliance on HTS methodologies. For example, a typical screen can cost several hundred thousand dollars to complete and the compound archive is systematically degraded.

The attractive potential of in-silico screening methodologies is therefore clear. The in-silico screening of an ~4000 compound "kinase directed" subset of the Novartis Pharma corporate archive was thus performed on a number of protein kinase targets using GOLD from CSD. These in silico HTS simulations are compared with empirical in-vitro HTS screening data on the same kinase targets and same compound dataset. The potential of this method as an alternative to in-vitro screening for hit identification is thus evaluated.

2:00 — 14. Fast, novel method of docking ligands to protein active sites. C. M. Venkatachalam⁽¹⁾, Jeff Jiang⁽¹⁾, Tom Oldfield⁽²⁾, and Marvin Waldman(¹⁾. (1) Molecular Simulations Inc, 9685 Scranton Road, San Diego, CA 92121, venkat@msi.com, (2) University of York, Molecular Simulations Inc

This talk will describe a fast method for docking ligands to protein active sites. The method consists of two parts as follows. First, using the 3-dimensional coordinates of the protein cavities are detected using a grid-based method and a flood-filling technique. The cavities can be edited and sorted according their sizes. A cavity is then chosen as a target for docking. In the second part, a Monte Carlo method is used for searching conformations of a given ligand. The shapes of the ligand and the cavity site are compared. The conformation is rejected if the shapes are not similar. If they are, the shape information is employed to immediately position and orient the ligand in the site.. The interaction energy of the ligand with the protein is estimated using grid-based energies. Softened van der Waals and electrostatic energies are obtained based on the potential suggested by M. Levitt. A novel non-linear interpolation scheme is employed which greatly reduces the errors in interpolating the van der Waals energy on the grid. The performance of this algorithm will be illustrated using several examples including HIV protease, thrombin, and DHFR.

2:30 — 15. Rapid combinatorial docking of chemical libraries. D. R. Lowis¹, S. Burkett², M. Snow³, and J. Bikker³. (1) Scientist, Tripos, Inc, 1699 South Hanley Road, St. Louis, MO 63144, Fax: 314-647-9241, dlowis@tripos.com, (2) Development, Tripos, Inc, (3) Parke Davis

Combined with appropriate scoring functions, docking has become a standard method for compound prioritisation in therapeutic programs. We present a novel approach to the efficient docking of combinatorial libraries. The method builds on the well respected FlexX program and the OptiSim selection algorithm. Docked solutions which closely mimic results obtained by FlexX itself are obtained in a much more timely and efficient manner by effectively docking reagent rather than product structures. Details of the method and results obtained will be presented.

3:00 — 16. Receptor-based virtual screening of very large chemical datasets. Carol A. Baxter, Jin Li, Christopher W. Murray, Tim D. J. Perkins, Richard A. Sykes, and Bohdan Waszkowycz, Protherics Molecular Design Ltd, Lyme Green Business Park, Macclesfield, Cheshire SK11 OJL, United Kingdom, Fax: +44 1625 500 666, carol.baxter@protherics.com

Slides (pdf)
Virtual screening involves the computational analysis of a large dataset in order to highlight those compounds most likely to be active. Protherics' virtual screening technology is based on the flexible-ligand docking package PRO_LEADS, which uses an empirical scoring function ChemScore to provide an estimate of binding affinity.

The DockCrunch project was designed to demonstrate the feasibility and advantages of performing virtual screening of a very large chemical database by receptor-based docking. The ACD-Screening Collection, which comprises over one million compounds, was docked against a 3D model of the estrogen hormone receptor. Analysis of the docking results revealed excellent discrimination between known ligands and random compounds, allowing the selection of subsets significantly enriched with active compounds. The docking of 1.1 million ligands took approximately 6 days when distributed across a 64-processor SGI Origin. On this time scale, virtual screening of very large libraries can become a routine task.

3:30 — 17. Design of combinatorial libraries for the rapid lead optimization of glucagon antagonists. Shenghua Shi¹, Atsuo Kuki¹, Anthony Lai Ling², Anthony Ling², Michael Plewe², and Javier Gonzalez³. (1) Computation Department, Agouron Pharmaceuticals, Alanex Division, 3550 General Atomics Court, San Diego, CA 92121-1194, shs@alanex.com, (2) Medicinal Chemistry, Agouron Pharmaceuticals, Alanex Division, (3) Combinatorial Chemistry Technology, Agouron Pharmaceuticals, Alanex Division

Slides (pdf)
Small molecule combinatorial libraries designed with proprietary software technology have significantly accelerated the process of lead optimization for potent glucagon antagonists. In the software, which is built upon an extensive chemistry knowledge base, each reactant is characterized according to classes of affinity-generating features, affinity elements (AE). Reactants can be selected either by similarity to the lead compounds or by diversity requirements in the AE chemistry space. We have applied two design strategies in the lead optimization libraries: 1) the reactants for one reaction component are selected based on the similarity to the lead while the reactants for the other components are diversified; 2) for the given reaction component a portion of reactants are selected based on the similarity and the rest of reactants are diversified. Libraries designed with these strategies quickly brought the potency from 3.1 mM to 59 nM and established certain acyl-hydrazone compounds as being potent glucagon antagonists.

Section B

Use of Toxicological Information in Drug Design

Grand Hyatt -- Constitution Ballroom A
R. W. Snyder, Organizer

1:00 — 18. Keynote Address: Application of toxicology databases in drug development. Joseph F. Contrera, and Edwin J. Matthews, FDA's Office of Testing and Research (OTR), Center for Drug Evaluation and Research (CDER), 5600 Fishers Lane, Rockville, MD 20857, Fax: 301-827-3787, contrerajf@cder.fda.gov

Slides (ppt)
In drug development, the application of combinatorial chemistry and high through put screening of compounds has resulted in an unprecedented increase in the number of compounds identified with potentially desirable pharmacological properties. The selection of lead compounds is currently hampered by limitations in the available methods for assessing toxicity. CDER files are a unique repository of the results of clinical and non-clinical toxicology studies. With the major advances in computer and information technology this unique scientific resource can be more effectively used to improve the scientific basis of regulatory decisions and product development. A current challenge is developing means to identify useful relationships and insights from large datasets.

Under a cooperative research and development agreement (CRADA), the FDA CDER modified and enhanced the capability of Multicase computational toxicology software to predict the carcinogenic potential of molecules based on chemical structure. Such software has potential regulatory and drug development applications that can ultimately benefit the public health.

1:30 — 19. Computational toxicology and virtual development in drug design. Dale E. Johnson, and Grushenka H. I. Wolfgang, ddplatform LLC, 6027 Christie Ave, Emeryville, CA 94608, djohnson@ddplatform.com

Slides (ppt)
The attrition rate of lead compounds that are optimized to development candidates and that eventually reach early clinical trials is still alarmingly high. One key reason for the high failure rate is undesired or unpredicted toxicity - either in animals or humans. Currently, a wide array of predictive tools are being developed with the goal of improving lead compound selection at the earliest stage. These new in silico approaches are changing toxicology into a knowledge- and information-based science where structure-toxicity-relationships are beginning to be elucidated. Predicting potential toxicity requires a knowledge of both ADME and pharmacokinetic data across species as well as an understanding of mechanisms of action and the chemical species that elicits the initial toxic event. Altough the toxicity "bottleneck" is formidable, virtual approaches using simulation algorithms are expected to accelerate and increase the success of early drug development within the next 3-5 years.

2:00 — 20. Paradigm shift from traditional to virtual. Stephen K. Durham, Lead Safety Assessment, Bristol-Myers Squibb, P.O. Box 4000, Princeton, NJ 08543, Fax: 609-252-7156, stephen.durham@bms.com

Slides (ppt)
The intensely competitive global pharmaceutical business environment has forced the need for the early and successful selection of viable drug candidates, and the abandonment of the traditional development paradigm. The enormous drug developmental costs necessitate the identification of toxicologic liabilities of novel compounds during the initial and least expensive phase of the discovery process. The incorporation of in silico programs to predict toxicity, accompanied by aggressive model development, will be fruitful avenues worth pursuit in the drug candidate evaluation algorithm.

2:30 — 21. Application of computational toxicology (ComTox) and multicase (MCASE) software to the FDA Mission. Edwin J. Matthews, and Joseph F. Contrera, FDA's Office of Testing and Research (OTR), Center for Drug Evaluation and Research (CDER), 5600 Fishers Lane, Rockville, MD 20857, Fax: 301-827-3787, matthewse@cder.fda.gov

Slides (ppt)
Under a CRADA between FDA and Multicase, Inc., new ComTox software programs have been developed that estimate chemical-toxic responses and dosages in animals using animal toxicology studies, and in humans using pre-market clinical-trial data and post-market adverse-event data for pharmaceuticals. This talk will examine two software modules that predict the potential carcinogenicity, and the maximum-tolerated-dose (MTD), in rats and mice, and it will elaborate on the experimental parameters that were accounted for the program's high predictive performance and excellent coverage for FDA-regulated substances. The parameters include: (1) large control data sets (n1000), (2) separate modules for each study cell ( / rat & mouse), (3) the use of biological potency scales, and (4) a human expert system. The talk will also review the current and future regulatory applications of ComTox within the Agency.

3:00 — 22. Data mining of toxic chemicals and database-based toxicity prediction. Jiansuo Wang, and Luhua Lai, Institute of Physical Chemistry, Peking University, Beijing 100871, China, Fax: 86-10-62751725, wjs@mdl.ipc.pku.edu.cn

Slides (ppt)
In the early stage of drug discovery, especially for computer-aided drug design,a large number of molecues will be proposed as potential leads and the bioactivity risk of these molecules is expected to be evaluated prior to synthesis. The rule-based expert systems have been used for the aim, while mining of a large amount of toxicological data can provide us with another promise.

In term of pharmacologists/toxicologists, toxicants are the drugs that cause vital harm. Therefore, the biochemical basis of toxic chemicals is the same as that of drugs and there exist toxicant-receptor systems just like drug-receptor systems. Under such a notion, we introduce some concepts and technologies, which are developled in drug design, into toxic chemicals, and conduct the following work.

I. We have studied the structural features of toxic chemicals from the RTECS database associating with specific toxicity. Potential active frameworks, groups and structure patterns for specific toxicity are obtained by computational chemistry approaches. These structural features of toxic chemicals will be helpful to understand activities of toxic chemicals and useful to predict toxicity of chemicals, especially in the early stage of drug design.

II. We take a two-step strategy to explore noncongeneric toxic chemicals from the database RTECS: the screening of structure patterns and the generation of detailed relationship between structure and activity. From the performance of overall procedure, such a stepwise scheme is demonstrated to be feasible and effective to mine a database of toxic chemicals.

III. We develope one programme as databased-based toxicity predictor of chemicals (dbToxPre). For one activity-query molecule, the programme firstly retrieves its structure-related molecule set by quick shape comparison with the molecules in toxicological database; then carries out detailed structure-toxicity-relation analysis to the molecule set and produce the toxicity prediciton of the query molecule. The programme mainly include four parts: a) a fast and efficient clustering of molecules based on molecular shape. b) field-based similarity computation of molecular structure based on shape cluster. c) flexible CoMFA analysis of molecules based on shape cluster. d) a database of toxic chemicals suitable for such procedure.

3:30 — 23. In silico toxicology screening of estrogenic compounds as potential therapeutic agents. William J. Welsh, Department of Chemistry, University of Missouri, 8001 Natural Bridge Road, St. Louis, MO 63121, Fax: 314-516-5342, WWELSH@UMSL.EDU

Specific estrogen-like compounds known as Selective Estrogen Receptor Modulators (SERMs) have attracted tremendous interest in the pharmaceutical industry and elsewhere as potential therapeutic agents for myriad applications in medicine. We have employed an integrated array of ligand-based and receptor-based approaches to design a large number of estrogen agonists and antagonists spanning several chemical classes. As part of this drug discovery program, we have developed and applied various in silico strategies for rapid screening of these compounds in terms of their toxicological and environmental effects. My talk will discuss growing concerns about the possible environmental impact of estrogenic compounds. It will also describe our computer-based models for predicting the biological activity and toxicological profile for the novel set of compounds under development in this laboratory.

MONDAY MORNING

Section A

8:25 — Introductory Remarks.

8:30 — 24. Working effectively in a global organization: Using technology to support customers. Joanne L. Witiak, and Thomas H. Pierce, Information Services Dept, Rohm and Haas Company, 727 Norristown Road, Spring House, PA 19477, Fax: 215-641-7857

Within the Rohm and Haas Company, the evolution of Information Services to the Knowledge Center has bolstered our work in the ongoing migration from physical information resources to those that can de delivered via a "virtual information center." When the company doubled its size though the merger with Morton Chemicals last year, it became even more important to network with and serve new employees throughout the global organization. We continually look for ways to effectively use technology to market our resources and services and educate and work with each other. We will discuss our experiences using Microsoft NetMeeting and web broadcasting and recording technology to deliver informational seminars, provide training and work with individuals and groups. Advantages include: the ability to provide support without the need to travel to remote sites; a broader network of employees; and the ability to deliver training to employees who can follow along at their desktop, in real time or asynchronously. Our challenge is to provide enough bandwidth within the corporation so that performance remains acceptable.

9:00 — 25. Marketing client databases: Opportunities for networking and instruction. Leah R. Solla, Physical Sciences Library, Cornell University, 293 Clark Hall, Cornell University, Ithaca, NY 14853-2501, Fax: 607-255-5288, lrm1@cornell.edu

Slides (ppt)
Dealing with the distribution and technical challenges of using client-databases in an academic institution provides opportunities to work directly with patrons and their research interests and needs. Specialized database clients are designed to provide very sophisticated information retrieval options. These specialized, sophisticated options can be tailored to the diverse nature of patrons information needs. Such rich databases as Beilstein, Gmelin, and Chemical Abstracts can be opened up to such diverse chemical users as food scientists, geologists, textile engineers. The few extra steps involved in accessing these client databases encourage a point of service. This is the opportunity for librarians to work more directly with patrons in their research space, to deliver customized and effective instruction on the functionality of these databases, and to network over research interests and live use of electronic resources. This presentation will focus on introducing Beilstein Commander and SciFinder Scholar at Cornell University over the past year.

9:30 — 26. Finding chemical information in the Pfizer research environment: SciFinder. Pamela J. Scott, Information Resources, Pfizer, Inc, Eastern Point Road, MS 118W-04, Groton, CT 06340, Fax: 860-715-7353, Pamela_J_Scott@groton.pfizer.com

Slides (ppt)
A non-vendor view of adminstering SciFinder in a corporate/ research environment. Beginning with a small pilot group, to global use of SciFinder by over 1,500 scientists. Issues to be discussed include new user orientation, education & training, web support, installations & upgrades, usage information, and new links to full-text & document delivery.

10:00 — 27. Catalyzing access to chemical information. George S. Porter, and Daniel Taylor, Caltech Library System, 1-43, Pasadena, CA 91125-4300, Fax: 626-431-2681, george@library.caltech.edu

Information seeking dynamics is analogous to thermodynamics. No chemist would expect a reaction to proceed rapidly if there are significant energy barriers. Catalysts are commonly sought, and employed, to create favorable energy costs for desired reactions. Chemical information specialists must realize the library’s role in catalyzing access to chemical information.

A suitably desirable product will be produced when the perceived value of the product is commensurate with the perceived costs of production. Libraries are the critical link, breaking down the access barriers through the provision of ubiquitous workstations configured to make the access as effortless as possible. Adoption of chemical information resources in the lab and office will occur when the value is perceived to outweigh the difficulties of the workstations in those locales.

The Caltech Library System has examined the many common file formats used for publishing and distributing chemical information via the Internet. Librarians have identified specific software (plug-ins and helper applications) necessary to support access to specific resources.

10:30 — 28. Developing the management of SciFinder within a global organization. Judith L. Johnson Philipsen¹, Martine Hanno², Jean-Gael Rouchon², Bernard Viratelle², Janet Liggett³, Pam Kubiak³, and William T. Turner¹. (1) Knowledge Management Services, Parke-Davis Pharmaceutical Research, 2800 Plymouth Road, Ann Arbor, MI 48105, Fax: 734-622-7008, Judith.Johnson@wl.com, (2) Service Information Scientifique et Communication, Institut de Recherche Jouveinal/Parke-Davis, (3) Library Services, Agouron Pharmaceuticals

Slides (ppt)
The Parke-Davis Research Library has been involved with SciFinder, since the alpha testing of its prototype. The enthusiasm for the launched product was such that Parke-Davis was the first company to purchase SciFinder in 1995 and since that time, the subscription has increased to include five countries and three states. To manage this global program, several support mechanisms have been developed. A network of support representatives, one from each site, through whom communications are channeled, has been established. Webpages, containing training summaries and other information, links to the CAS SciFinder page, and access to program downloads, have been created at several sites. Training is provided not only in seminars by CAS representatives but in smaller groups or individually by the support network members. The use profile of the scientist is assessed so that the type of seat assignment is most cost effective. The effects of SciFinder upon the end-user access to chemical information and current awareness, as well as on the change in search requests for the professional searcher, will be discussed.

11:00 — 29. Dehorning the client/server dilemma. F. Bartow Culp, and Song Yu, Libraries, Purdue University, Mellon Library of Chemistry, 1538 Wetherill, West Lafayette, IN 47907-1538, Fax: 765-494-1579, bculp@purdue.edu

The delivery of electronic resources directly to the chemist's desktop via web- and client/server-based systems is rapidly dissolving the physical limitations of the classical library and changing the role of the librarian. However, the administration of these services presents the librarian with a dilemma: Widespread local distribution of these systems can create a logistical nightmare, as when they need to be upgraded; however, limiting distribution to a few workstations can artificially restrict their use. The Purdue University Libraries has implemented a system using Windows Terminal Server (WTS) from Microsoft and Metaframe from Citrix Systems, Inc. It allows users, by downloading a single client, to gain access to the current versions of a wide variety of client-based resources such as SciFinder Scholar and Beilstein CrossFire, which are locally mounted on the Libraries server. Valid users of these and other resources can gain access to them without geographic restrictions, and can print and save files locally. We will demonstrate how the Purdue Libraries WTS system works, and discuss some of its strengths and limitations.

11:30 — Discussion.

Section B

8:55 — Introductory Remarks.

9:00 — 30. Academic choice. Patricia E. O'Neill, Pacific Lutheran University, Mortvedt Library, Tacoma, WA 98447, Fax: 253-535-7315, oneill2@plu.edu

Slides (ppt)
A library degree and a good solid chemistry background. Industrial positions are available at nice salaries with good benefits. Why choose a lower paying job to work in an academic library? There are many benefits to the academic choice. One is the traveling pension fund. Another the focus on education. Choosing to focus your career in the academic world is only the beginning. Academic job ads show there are many more choices to be made. Should you work in a large multidisciplinary library? Should you find a small university and be their science librarian? Is it important to be at a Research Library Group university? Maybe a specialist position is the right way to go. Then come the questions about atmosphere and titles. Is being a tenured faculty member important? How much additional work will that take? What about tenure-like positions? Do they have the same respect? The choice to work in academia is only the beginning. Within the academic circles there are many alternatives. Knowing what the options are is the first step. As your career progresses you may find you need to reassess those choices and consider new options. Personal experience at Emory University, Cornell University, and Pacific Lutheran University allow me to take advantage of the flexibility inherent in an academic career.

9:30 — 31. Horizontal thinkers wanted: Chemical information science from the industrial perspective. Steve Boyle, Library & Information Services, Nalco Chemical Company, One Nalco Center, Naperville, IL 60563-1198, Fax: 630-305-2876, sboyle@nalco.com

Industry offers rewarding and sometimes exciting career opportunities in chemical information science. Market sectors with a high demand for information support include pharmaceuticals, biotechnology and specialty chemicals. Individual background, skills, interests and personality all contribute to the direction and ultimate success of an industrial chemical information career. Also important are environmental factors such as where R&D and information support fit within the larger organization. This paper will discuss information science within the traditional career development model of self-assessment, investigation, matching and career planning. Examples from the petroleum and specialty chemical industries will be used to illustrate how experience in the industrial realm compares to that in academic or other non-industrial assignments. Finally, some current trends in information services will be highlighted, along with their effects upon the career planning and management process.

10:00 — 32. Chemical informatics careers at NIST: Reliable data for chemistry. John Rumble Jr., Standard Reference Data, NIST, 100 Bureau Drive MS 2310, Gaithersburg, MD 20899-2310, Fax: 301 926 0416, john.rumble@nist.gov

The National Institute of Standards and Technology (NIST) operates the Standard Reference Data Program, one of the largest and most diverse data programs in the world. Today it has data evaluation and data dissemination activities in about 40 scientific disciplines, many involving chemical data. Chemists and others work to collect, evaluate and disseminate data using the most advanced methodologies available. NIST data products include the NIST Chemistry WebBook, the NIST Mass Spectral Library, the Protein Data Bank and over 15 other chemical databases. Among the data related careers at NIST are four of prime importance: data evaluator, database designer and reviewer, online data systems developer and data project leader. NIST traditionally uses scientists in all these positions, even for most programming activities. I will describe these different positions, as well as how people get started in them, and identify the skills and background including scientific, programming, management, teamwork, and communication.

10:30 — 33. Non-traditional careers in chemistry at the American Chemical Society. Lorrin R. Garson, American Chemical Society, 1155 16th Street NW, Washington, DC 20036, lrg96@acs.org, and Robert L. Swann, Chemical Abstracts Service

Slides (ppt)
The American Chemical Society as an organization offers an alternative, non-traditional career path for chemists--largely in chemical information and information technology. Chemical Abstracts Service employs chemists in indexing and abstracting roles and the Publications Division employs chemists in technical editing. Both divisions employ chemists and other scientists with computer skills in the development and support of sophisticated information systems. The roles of scientists at the ACS will be discussed in detail.

11:00 — 34. Career opportunities at CAS. Karen C. Toplek¹, D. Jeanne Jones², and Deborah Near². (1) Molecular Biology and Genetics, Chemical Abstracts Service, 2540 Olentangy River Rd., Columbus, OH 43202-1505, Fax: 614-447-3713, kct53@CAS.org, (2) Biochemistry, Chemical Abstracts Service

As access to information has become an increasingly important competitive advantage around the globe, Chemical Abstracts Service (CAS), a division of the American Chemical Society, has gained recognition as the world's largest storehouse and leading provider of chemical and biosequence information. It is the only database of it's kind to offer both chemistry-related literature and patents.

CAS professionals are meeting the challenges of the information explosion by working together in a variety of departments across CAS, including Editorial, Research, Information Systems, New Product Development, and Marketing.

Individuals who excel in scientific and technology-rich environments and enjoy applying their state-of-the-art knowledge, innovation and experience in the workplace will find exceptional career opportunities within these divisions at CAS. Plan to attend this seminar to learn the educational requirements and job skills necessary to become part of this world-renowned team that is facilitating future scientific research through premier chemical information technology.

MONDAY AFTERNOON

Section A

1:00 — Introductory Remarks.

1:05 — 35. Developing and implementing a formulations and chemistry decision support system. Michael J. Doyle, MSI, 3483 Camino Michelle, Carlsbad, CA 92009, Fax: 858-799-0136, mdoyle@msi.com

Formulation sounds easy. Take well-known ingredients and mix them to make something with better properties or performance. Make a tablet that's stable under air contact and room temperature. Make a cough mixture that costs less than existing formulas, but still tastes great. Deliver a new drug in a comfortable fashion that delivers the same dose every time, with few side effects. If this still seems straightforward, did your tablet just get too large for patient comfort? Or if your intermediate materials get stuck in a railway yard are they likely to turn to concrete? Can you make varieties of your anti-asthma mixture to match preferences in different continents? A number of trends in the pharmaceuticals industry magnify these problems. Leading companies run global operations in global markets, so formulators must respond to diverse customer-bases and collaborate with colleagues worldwide. Intense competitive pressures mean that more products must be generated faster. Trial-and-error is becoming an even less secure basis for competitive advantage. This role of accelerating and supporting the complex task of pharmaceutical formulation can be performed using software tools. Creating such a system needs existing technology to be creatively applied and extended. Its applications will cut across a wide range of products and industries, and provide significant competitive advantages to early-adopters, yet core requirements for the system can be established with no sharing of proprietary information. It calls for a synthesis of software, systems, and scientific skills not available in many companies. These are the ideal ingredients for a pre-competitive consortium. The industry-academic-software vendor consortium approach will help to accelerate and determine the best way to combine and implement technologies to maximize their impact on these commercially significant problems. Member organizations typically join for three-year periods. Membership entitles researchers to use the consortium software, provide input to software development, attend regular meetings where ideas are shared with peers, receive dedicated application support, and gain early access to new technology.

This talk will cover the aspects of the rise of the consortium approach as well as discussing the evolution of software modeling into decision support tools in a client server environment.

1:35 — 36. Garbage in - results out: Name normalization in ChemFinder.com searching. Louis J. Culot Jr., and Ni Yan, Informatics, CambridgeSoft Corporation, 100 Cambridge Park Drive, Cambridge, MA 02215, lculot@camsoft.com

Despite a large scientific audience, many of ChemFinder.Com's search requests are ill formed, and would not match standard chemical names / synonyms. Text matching, even against exhaustive synonym lists, will not yield matches using standard text-matching algorithms. While the computer is unforgiving, most scientists can readily defer the intended query and intelligently suggest results. Through the use of name and query normalization, CambridgeSoft has dramatically improved the quality of search results for poorly-formed queries. By pre-processing the stored chemical names, while preserving the full molecular meaning, and by likewise normalizing the queries, we were able to yield a significant increase in the success rate.

2:05 — 37. Merging disparate chemical information sources. Louis J. Culot Jr., and Irwin Schreiman, Informatics, CambridgeSoft Corporation, 100 Cambridge Park Drive, Cambridge, MA 02215, lculot@camsoft.com

Chemical information exists in a variety of formats and sources: Robust and well-checked databases, small group databases, Excel spreadsheets and Word documents, and so forth. In building the ChemACX database (a comprehensive chemical sourcing database), we encountered virtually every sort and quality of chemical information. Two approaches are commonly used in managing disparate data sources: merging the data, and building data-searching applications which combine queries and results. We explore the applications for both approaches, and focus on good practices for merging data. By applying consistent rules, using data sources as quality controls, and developing an algorithm for merging the data, the resultant database was better checked and of higher quality than the input data, and was built in a highly-automated way. We also maintain a high degree of original data integrity, and provide mechanisms for updates and distribution.

2:35 — 38. ACS Library and Information Center: Evolving from a collection to a service. Svetla P. Baykoucheva, ACS Library and Information Center, American Chemical Society, 1155 Sixteenth St., NW, Washington, DC 20036, Fax: 202-872-6257, s_baykouchev@acs.org

The role of the ACS Library and Information Center (previously, ACS Library Services) has dramatically changed. In addition to performing all typical library duties, the ACS Library is now offering many new services. These include a New Digital Imaging Center; publishing the ACS newsletter; staff training; serving the public through the Chemical Health & Safety Referral Service; providing a content-rich chemical information on the ACS Web page (http://www.acs.org/infocenter.html); and participating in the speakers service for local ACS sections. The challenges of imposing a new model to a traditional special library - and of making it work - will be discussed.

3:05 — 39. Chemistry 2000: The first 2000 days. Jonathan M. Goodman, Department of Chemistry, Cambridge University, Lensfield Road, Cambridge CB2 1EW, United Kingdom, Fax: +44 1223 336362, jmg11@cam.ac.uk

Slides (htm)
A WWW index of university chemistry departments and chemistry journals has been maintained in Cambridge for more than five years. (http://www.ch.cam.ac.uk/ChemSitesIndex.html). This is not the only such list available world wide (see, for example: On the need for multiple lists of chemical information J M Goodman Molecules, 2000, 5, 33-36), but it has a distinctive flavour. The content of the list changes quite rapidly, and would quickly become obsolete without a strategy for continual updating. This task must be automated so far as possible in order for it to be manageable. Programs have been developed to do this, which make the maintenance process more than 95 % automatic. The programs which are used to do this and strategies for future development will be discussed.

3:35 — Discussion.

Section B

1:25 — 40. Careers in patents. Barbara Burg, Information Services, DuPont Pharmaceuticals, Chestnut Run Plaza, MR-1276, Wilmington, DE 19805, Fax: (302) 892-1686, barbara.burg@dupontpharma.com

An alternative to lab work is a career in Patents. Various levels of patent positions in industry, private practice and the Patent Office will be discussed. Some of the questions to be answered are: What level of education is required? Are positions available in my area of science or technology? What type of work is involved? And no, you don't need to be a Patent Agent or Attorney to find a position in patents, but more avenues open up if you do.

(htm slides)
1:55 — 41. Making the transition from special to academic librarianship. Philip Barnett, Science/Engineering Library, City College of New York, Convent Avenue at 138th Street, New York, NY 10031, Fax: 212-650-7626, phbcc@cunyvm.cuny.edu

The transition from working in either a special library or corporate information center to an academic library requires some adjustments. The least adjustment is the change from chemical information specialist to chemistry librarian in a dedicated chemistry library primarily used by advanced chemistry students and chemistry faculty. Most librarians, however, must either serve in multidisciplinary science libraries, or become science specialists in single campus libraries. In such situations librarians may face new responsibilities. Generally, librarians teach more in academic libraries, often to novice library users. They may teach credit bearing chemical information courses. Some academic librarian positions carry faculty rank with its accompanying perks, such as tenure and generous vacation, and accompanying responsibilities, such as the requirement to publish and perform service to the greater library community. Most librarians can be content in both academia or industry.

2:25 — 42. Unstructured career in chemical information. Charles W. McFarland, Research Dept, McGean Inc, 2910 Harvard Ave., Cleveland, OH 44105-3010, Fax: 216-441-1377, cwm001@apk.net

It is possible, especially in the context of a smaller company, for a career to gain a substantial chemical information component without particular formal training in this specialty. Such a situation is the framework for this talk, which is primarily a personal, anecdotal history, from an undergraduate introduction to the classic print references in the early sixties, to the use of the Internet in the nineties. This account in part will show how the value of information can be substantial, such as a single Usenet posting leading to the formation of a new corporate division. "If you can think of it, odds are someone has written about it" and other individual viewpoints on what facilitates successful information retrieval will also be given. Although the experiences related are perhaps most relevant to those contemplating career changes, the insights gained should also benefit those just entering the chemical profession.

2:55 — 43. Options for a chemist in the information field. Randall K. Ward, Information Services, DuPont Pharmaceuticals Company, MR1136, POBox 80721, Wilmington, DE 19880, Fax: 302-992-2821, randall.k.ward@dupontpharma.com

Slides(ppt)
Some chemists are quite content in the laboratory and wish to make a career there. Others may not be as content. There are different kinds of people, with different likes and dislikes. All kinds of people are needed in a successful enterprise. As a bench chemist, what are some options and potentials in the information field? This presentation will attempt to describe some of them, based on the experiences of one who worked in an organic synthesis lab for nine years and then transferred to information (formerly known as "library") services. Topics covered include suggestions for making the transition from lab chemistry to a new field, what one will find that's different about the information field, and the strengths a chemist brings to the table.

3:25 — Panel Discussion.

Open Meeting -- Committee on Chemical Abstracts Service and ACS Publications

ACS Publications PowerPoint presentation

MONDAY EVENING

44. 'Leadlessly' leading the way: Wireless Science. Bryan A Vickery, ChemWeb Inc, 84 Theobald's Road, London WC1X 8RR, United Kingdom, Fax: +44 (0) 20 7611 4301, bryan.vickery@chemweb.com

Abstract text not available.

45. Construction and application of molecular equivalence number. Yong-jin Xu, and Mark Johnson, Computer-Aided Drug Discovery, Pharmacia Corporation, 301 Henrietta St., Kalamazoo, MI 49001, Fax: 616-833-9183, Yong-Jin-Cons.Xu@am.pnu.com

Assigning a common molecular equivalence number to compounds with the same chemical feature, for example the same cyclic system, enables a linear arrangement of compounds in a database so that the compounds with the same structure feature are consecutively positioned. The chemical graph has long been one of the fundamental representations of molecular structure. Therefore, mapping the chemical graph, or more generally, a labeled pseudograph, to a unique number is the key step for constructing the molecular equivalence number. Computing this number via the Morgan algorithm, we observed a large classification noise in which non-isomorphic graphs were mapped to the same number. Our extensions to Morgan algorithm greatly reduce the classification noise for most chemical graphs. A variety of interesting molecular equivalence relations can be defined and the corresponding molecular equivalence numbers computed. The list of all or some of these numbers, whose selection and order depends on one's chemical interest, can be used as a string-valued molecular descriptor. The descriptor, when lexicographically ordered, facilitates the sorting, grouping and comparing the chemical database in a particular chemical sense.

46. Undergraduate assignments in Beilstein and Gmelin CrossFire. Ann D. Bolek, Science-Technology Library, University of Akron, Akron, OH 44325-3907, Fax: 330-972-7033, bolek@uakron.edu

During their junior year, undergraduate students in Advanced Chemistry Labs are given three assignments in Beilstein and Gmelin CrossFire. They search for particular properties by chemical name in the first assignment, by chemical structure in the second assignment, and for chemical reactions in the third assignment. This poster will illustrate the three assignments.

47. Wahoo! A corporate portal. Mary Moulton, Venkata Balabhadra, Deborah Einhorn, Stacy Hortner, and Karen Tatarowicz, Information & Library Services, Air Products & Chemicals Inc, 7201 Hamilton Boulevard, Allentown, PA 18195-6517, Fax: 610-481-6517, moultome@apci.com

Although it is a valuable source for useful information, the world wide web is not a library. Internet resources seldom have an editor or publisher responsible for content; organization and standards for accuracy do not exist. Attempts to make organized lists of reliable internet resources available on the company intranet proved to be labor intensive and inefficient. Wahoo! was developed to improve access to internet resources, eliminate duplication of efforts, distribute maintenance responsibilities and take advantage of the expert knowledge of Air Products employees. Wahoo! is a searchable database of selected, evaluated, hierarchically organized links to external web sites. It is accessible from the company intranet. Individuals having an interest in specific technical or business areas volunteer to maintain subject categories which they populate with annotated and reviewed links.

48. Publishing scientific data for electronic books: Challenges and opportunities. Geraldine Dalton, Standard Reference Data, National Institute of Standards and Technology, 100 Bureau Drive, Stop 2310, Gaithersburg, MD 20899-2310, dalton@nist.gov

Slides (ppt)
The Standard Reference Data Program at the National Institute of Standards and Technology (NIST) has recently initiated a project that will provide our customers with the opportunity to obtain our data products in electronic book format (ebook). The ebook offers yet another possible market for NIST data products for applications where data is required but access to web or PC based systems is restricted. More importantly, offering the ebook format adds almost no additional cost to the data since the format is similar to the on-line product and utilizes an open version of XML. This presentation will provide an overview of our efforts in the emerging area of ebooks and offer some thoughts on the impact of this technology to the chemistry community.

TUESDAY MORNING

Section A

Virtual High-Throughput Screening
ADME property-based Screening

Grand Hyatt -- Constitution Ballroom A
Cosponsored with Division of Computers in Chemistry, and Division of Medicinal Chemistry
O. F. Güner, Organizer, Presiding

8:55 — Introductory Remarks.

9:00 — 49. Integrated approach to in silico screening. Joseph L. Durant Jr., Douglas R. Henry, Maurizio Bronzetti, and David A. Evans, MDL Information Systems, 14600 Catalina Street, San Leandro, CA 94577, jdurant@mdli.com

Slides (pdf)
As the cost of pharmaceutical R&D continues to increase so does the requirement to find ways to screen compounds earlier and more reliably - the so-called fail early paradigm. One of the strengths of computational chemistry is the ability to take huge numbers of compounds - a virtual screening library - and perform rigorous 'in silico' screening tasks. One of the main difficulties in this process is the ability to collect, manage and analyze data from a variety of different sources. We present an integrated informatics framework that enables a variety of independent tools to be used to design, screen and re-design virtual libraries. With a small targeted library example we show the necessary components of these cycles including molecular fingerprinting, descriptor calculation, data visualization, ADME and toxicology prediction and searching.

9:30 — 50. Informatics challenges in data management, high-throughput screening and development of predictive models of ADME properties. Peter Gund¹, Janet Cohen¹, William J. Egan¹, Osman F. Güner², and Kirk McMillan¹. (1) Pharmacopeia, Inc, 3000 East Park Blvd., Cranbury, NJ 08512, pgund@pharmacop.com, (2) Molecular Simulations Inc

Slides (pdf)
While practically every pharmaceutical and chemical research organization has preclinical chemical information management systems, ADME/Tox data are usually held in restricted access preclinical and clinical information systems, due to the sensitivity of much pharmacokinetic and toxicity data. Although it is now generally recognized that these data are extremely useful in the early stages of drug discovery projects, commercial information systems do not enable the practicing medicinal chemist to query ADME/Tox data along with structures and calculated properties. In this presentation we describe the design of such an ADME/Tox Information System, we highlight some high-throughput ADME screens run at Pharmacopeia, and we indicate how they have been used to develop predictive models of pharmacokinetic properties of drug leads.

10:00 — 51. Virtual high-throughput screening based on bioactivity datamining of CAS databases. W. Fisanick, G. S. Ethiraj, R. J. Schenck, and Q. Yuan, Research and New Product Development, Chemical Abstract Service, 2540 Olentangy River Road, P. O. Box 3012, Columbus, OH 43210, Fax: 614-447-3813, wfisanick@cas.org

Chemical Abstracts Service (CAS) has developed datamining techniques for browsing and predicting substance bioactivity using data in the CAS Registry and CAplus databases. The knowledge is derived from the implicit relationships created by CAS scientists over years of abstracting and indexing the world’s journal and patent literature. Scoring methods allow setting thresholds for the processing of virtual libraries. Qualitative estimation for specific therapeutic categories, e.g. anticonvulsant, antihypertensive, etc., specific modes of action, e.g. beta blocker, and specific disease relationships are made for each member in the virtual library. In addition, a general "drug-likeness" value is calculated for each substance as an aid to focussing the diversity of the library on substances with a good bioactivity potential. Other considerations are whether a library substance has been added to the CAS Registry, appeared in a patent, or has a commercially-available synthetic precursor, etc. This paper will discuss screening results for some example substance sets.

10:30 — 52. Using molecular structure to assess permeability. Stephen R. Johnson¹, Fan Lin², HY Cheng², Sunny Hung¹, Joel Saunders¹, Weifan Zheng¹, and George L. Seibel¹. (1) Cheminformatics Department, SmithKline Beecham Pharmaceuticals, 709 Swedeland Road, King of Prussia, PA 19406, Stephen_2_Johnson@sbphrd.com, (2) Physical and Structural Chemistry, SmithKline Beecham Pharmaceuticals

In the last several years, consideration of the oral bioavailability of drugs has assumed greater importance in the early discovery phase of pharmaceutical research. The ensuing search for higher-throughput methods of evaluating intestinal permeability has resulted in methods capable of generating data in quantities sufficent for the development of reasonable structure-permeability models for use in library design, template selection, and prioritization of compounds. We will present a permeability classifier developed using data from an artificial membrane permeability assay. This model uses learning vector quantization pattern recognition to classify virtual compounds as either highly or barely permeable based on structural features. The accuracy of classification is sufficient for use in a discovery setting

11:00 — 53. Web-based computational tool for combinatorial library design that simultaneously optimizes multiple properties. Weifan Zheng, Sunny T. Hung, Joel T. Saunders, Stephen R. Johnson, and George L. Seibel, Cheminformatics Department, SmithKline Beecham Pharmaceuticals, 709 Swedeland Road, King of Prussia, PA 19406, weifan_2_zheng@sbphrd.com

Slides (ppt)
Combinatorial library design is by nature a multicriterion design problem. These criteria often include diversity, similarity and product novelty with respect to a corporate database. In addition to these structural properties, other physicochemical properties also need to be considered. To address this design problem, we have developed a computer program (PICCOLO) that can simultaneously optimize library properties such as diversity, similarity, product novelty, Lipinski properties, solubility, permeability, p450 inhibition and other ADMET properties. In this talk, we will describe the overall design of this program and the formulation of individual penalty scores. Some practical issues will also be discussed.

Section B

Skolnik Award Symposium -The Future of Chemical Information

Grand Hyatt -- Constitution Ballroom B
S. Heller and G. W. A. Milne, Organizer

8:50 — Introductory Remarks.

9:00 — 54. 3-D Pharmacophore fingerprints for virtual screening, library design and profiling. Jonathan S. Mason, Brett R. Beno, Daniel L. Cheney, Sung Jin Cho, and Andrew C. Good, Structural Biology and Modeling, Bristol-Myers Squibb, P.O. Box 4000, Princeton, NJ 08543, Fax: 609-252-6030, masonj@bms.com

The opportunities and challenges arising from combinatorial chemistry, virtual screening and high throughput screening have led to the development of new methods to exploit information that can be generated from both ligands and targets, and that approach the problem of "drug-likeness" and bioavailability . A ligand and target based method based on 3-D pharmacophore fingerprints (multiple 3D 3- and 4-point potential pharmacophores) for applications such as virtual screening, library design and profiling will be presented. Both ligand and protein-site information (shape and pharmacophoric) can be used; diversity or similarity can be measured and compared relative to a feature of interest or to a protein site in addition to a global or absolute sense.

9:40 — 57. Large chemical databases on the Web: Enhanced CACTVS Browser of the open NCI database. Marc C. Nicklaus¹, Frank Oellien², and Wolf-Dietrich Ihlenfeldt². (1) Laboratory of Medicinal Chemistry, Division of Basic Sciences, National Cancer Institute, National Institutes of Health, 37 Convent Drive, MSC-4255, Bldg. 37, Room 5B29, Bethesda, MD 20892-4255, Fax: 301-402-2275, mn1@helix.nih.gov, (2) Computer Chemistry Center, Institute of Organic Chemistry, University of Erlangen-Nuremberg

Since 1998, the Erlangen/Bethesda Data Services, a collaboration of the Computer Chemistry Center, University of Erlangen-Nuernberg, Germany, and the Computational Chemistry Section, Laboratory of Medicinal Chemistry (NCI, NIH), Bethesda, MD, USA, have made available structures and bioactivity data from the open part of the NCI database in searchable format through a web interface based on the CACTVS chemical information toolkit. This service, the "Enhanced NCI Database Browser," built from publicly available structural data of NCI's Developmental Therapeutics Program augmented by additional information, currently comprises more than 246,000 structures. They are searchable by various criteria, such as full structure, substructure, molecular similarity, test data etc. Hits can be displayed and downloaded in various formats, including an Image Gallery and web-specific formats such as VRML. We present an overview, and report on the new developments of this service, such as the addition of 3D (pharmacophore) search capabilities, additional searchable data associated with the compounds, and new links to other services.

10:20 — 55. Art of plotting structures and their data. Wolf D. Ihlenfeldt, Computer Chemistry Center, Institute of Organic Chemistry, University of Erlangen-Nuremberg, Nägelsbachstr. 25, Erlangen D-91052, Germany, Fax: +49 9131 852 6579, wdi@ccc.chemie.uni-erlangen.de

Slides (ppt)
The plotting of chemical structures with automatically generated 2D display coordinates is a complex problem if a comprehensive solution is desired. Complications such as crowded ring joints, folding of large rings and especially bond stereohemistry in such rings require algorithms which go beyond what is available in standard 'structure cleanup' algorithms.

Further problems arise when not just isolated compounds but rather groups of similar compounds are involved. Normalization of plot orientation with the aim to highlight similarities within a collection of compounds demand coordinate modification methods which detect common patterns and modify individual coordinate sets to conform to a virtual common framework, without degrading the visual quality of the individual plots.

We present a set of algorithms for structure layout which can handle the aforementioned problems. They are augmented by novel methodology for the visual analysis of compound similarity and compound property display, which also will be presented.

11:00 — 56. Harness the power of computer in drug discovery and design. Shaomeng Wang, Drug Discovery Program, Lombadi Cancer Center and GICCS, Departments of Oncology and Neuroscience, Georgetown University Medical Center, The Research Building, Room WP05, 3900 Reservoir Road, Washington, DC 20007, Fax: 202-687-0617, wangs@giccs.georgetown.edu

Over the last 30 years, computer in drug discovery has grown from its infancy to a mature age. Computer now plays an important role in drug discovery in the areas of chemical and biological information management and mining, QSAR, lead discovery and lead optimization, protein structure modeling, molecular docking and structure-based drug design. I will present in this paper our success in computer assisted drug lead discovery and lead optimization using ligand-based, structure-based and informtics-based approaches. I will also discuss our challenges and incredible opportunities we have in the post-genomic era using computer in drug discovery.

TUESDAY AFTERNOON

2:00 — Introductory Remarks

2:10 — 58. Chemical publication: A critical evaluation. Steven M. Bachrach, Department of Chemistry, Trinity University, 715 Stadium Drive, San Antonio, TX 78212, Fax: 210-999-7569, sbachrach@trinity.edu

Slides (htm)
Chemists communicate their science principally via written works, usually articles and communications printed in journals. The printed journal is now over three centuries old and the advent of electronic publishing demands a re-evaluation of how we communicate. In this talk, I will discuss publications trends, the role of the journal and publisher, and new technologies that directly bear on scientific communication. Whatever change might be useful or necessary are predicated on the active participation of authors, and the first step is educating the community. This talk will serve to motivate a community-wide discussion of the future of chemical publication.

2:50 — 59. NIST Mass Spectral Database through the ages. J. K. Klassen, Physical and Chemical Properties Division, National Institute of Standards and Technology, 100 Bureau Dr Stop 8380, Gaithersburg, MD 20899-8380, Fax: 301-869-4020, jane.klassen@nist.gov

The NIST/EPA/NIH Mass Spectral Library has mirrored the development of chemical information for the last three decades. The Library started as a remotely accessed database available for searching via an international time-sharing computer network. By 1978, the Library was additionally delivered as a 4-volume book set and as a magnetic tape installed on a variety of commercial GC/MS minicomputer systems. With the development of PCs, many improvements have been made to the database including translation from the magnetic tapes to PC format, additional data (e.g. chemical structures), advanced search algorithms, and a more complex user interface. The current Library contains fully evaluated mass spectral information for over 100,000 compounds. Future directions for the NIST Mass Spectral Database include adding a Retention Index database and estimator for added confidence in library searches and optimizing library search algorithms for ion trap spectra.

3:30 — 89. Wireless access to chemical information. Bryan A Vickery, ChemWeb Inc, 84 Theobald's Road, London WC1X 8RR, United Kingdom, Fax: +44 (0) 20 7611 4301, bryan.vickery@chemweb.com

WAP! - the world's two fastest-growing technologies - wireless communications and the Internet - are about to converge.

WAP (Wireless Application Protocol) is a communication protocol between a mobile terminal and an IT system which provides mobile access to services and content. The way will be lead by B2C (business-to-consumer ) commerce, through banking, ticket buying, traffic details and news. This time, however information providers and B2B (business-to-business) companies will not be far behind.

This talk will explore services "first off the mark" for chemistry and the chemicals industry and examine the future possibilities for delivering information to academics, researchers and business professional wirelessly - wherever you are!

4:10 — 90. Data -- Banks, bases, and formats, past present and future. The good, the bad, and the ugly. A. N. Davies. , SAS, Institute of Spectrochemistry and Applied Spectroscopy, Bunsen-Kirchhoff-Str.11, Postfach 10 13 52, 44013 Dortmund,Germany, davies@isas-dortmund.de

The ability to communicate with one another, exchange abstract thought and pass on communal knowledge between generations is supposed to have lead to the human race dominating the animal world. Having worked in the field of scientific data base building, data exchange standards and QC for the last 12 years I have marvelled at the tenacity of the crusaders who continuously strive to improve the general scientific social condition. These heroes have led a long hard struggle to bring scientific data communication out of the dark-ages. Some of the successes and spectacular failures of recent years will be highlighted and analysed. A brief introduction to the issues behind the acronym-wars between JCAMP-DX, IUPAC, NetCDF, AIA, ANDI, etc. The state-of-the-art will be sketched including information of the current players in this difficult field and emphasising open systems, databases and code. An attempt to see into the future will be dared!

WEDNESDAY MORNING

8:55 — Introductory Remarks.

9:00 — 60. Widely applicable set of 2½ D descriptors. Paul LaBute, Chemical Computing Group, Inc, 1255 University St., Suite 1600, Montreal, QC H3B 3X3, Canada, Fax: (514) 874-9538

Three sets of molecular descriptors computable from connection table information are defined. These descriptors are based upon atomic contributions to van der Waals surface area, log P (octanol/water), molar refractivity and partial charge. The descriptors are applied to the construction of QSAR/QSPR models for boiling point, vapor pressure, free energy of solvation in water, solubility in water, Thrombin/Trypsin/Factor Xa activity, blood-brain barrier permeability and compound classification. The wide applicability of these descriptors suggest uses in QSAR/QSPR, combinatorial library design and molecular diversity work.

9:30 — 61. Novel method of building regression tree models for QSAR based on artificial ants. Sergei Izrailev, and Dimitris K. Agrafiotis, 3-Dimensional Pharmaceuticals, Inc, 665 Stockton Drive, Suite 104, Exton, PA 19341, Fax: 610-458-8249, sergei@3dp.com

Among the multitude of learning algorithms that can be employed for deriving quantitative structure-activity relationships (QSAR), regression trees have the advantage of being able to handle large data sets, dynamically perform the key feature selection, and yield readily interpretable models. A conventional method of building a regression tree model is recursive partitioning, a fast greedy algorithm that works well in many, but not all cases. This work introduces novel methods of building regression tree models based on simulated annealing and artificial ants. These methods are shown to yield better models on some data sets than the models obtained by recursive partitioning.

10:00 — 62. High-throughput screening of virtual combinatorial libraries with neural networks. Victor S. Lobanov, 3-Dimensional Pharmaceuticals, Inc, 665 Stockton Dr., Suite 104, Exton, PA 19341, Fax: 610-458-8249, lobanov@3dp.com

According to the rules of combinatorics there are an astronomically large numbers of compounds that can be made available by means of the combinatorial chemistry. These numbers make even virtual screening of all possible compounds prohibitive if each structure has to be enumerated and evaluated for similarity or particular properties. Despite their huge number, structural diversity of the combinatorial products steams from a relatively small number of reagents. This feature of combinatorial libraries can be used to our advantage when virtual combinatorial libraries are subjected to virtual screening. This talk will present a new approach that allows dramatically increase throughput of in silico screening of virtual libraries by evaluating properties of combinatorial products based on the features of the corresponding reagents by means of neural networks. We will present examples of similarity and property based selections generated by the new technique.

10:30 — 63. Molecular docking: A pharmacophoric approach. Bert E. Thomas IV, Diane Joseph-McCarthy, Michael Belmarsh, Demitri Moustakas, and Juan C. Alvarez, Biological Chemistry, Wyeth Research, 87 CambridgePark Drive, Cambridge, MA 02140, Fax: 617-665-8993, bthomas@genetics.com

The rapid growth of small molecule databases available for virtual screening, both commercial and proprietary, necessitates that molecular docking algorithms should be minimally able to handle hundreds of thousands of molecules, thus placing an upper limit for evaluation of seconds, rather than minutes, on a per molecule basis. One approach aimed at satisfying this requirement is to address conformational flexibility of the putative ligands, which is necessary for accurate screening, at the database generation step, not during the docking procedure. Docking each conformer individually would be computationally prohibitive. Our solution to this problem is to generate pharmacophore-based ensembles of conformers. The pharmacophore of each ensemble is docked and the associated conformers are then scored for electrostatic and shape complementarity. This new virtual screening method will be described.

11:00 — 64. Classification of biological binding sites for large scale virtual screening of combinatorial libraries and design of general purpose organic scaffolds. James R. Arnold¹, Michelle L. Lamb¹, Scott C-H. Pegg¹, Keith W. Burdick², Samuel Toba¹, Thomas A. Robertson¹, R. Kip Guy¹, and Irwin D. Kuntz¹. (1) Department of Pharmaceutical Chemistry, University of California, 533 Parnassus Avenue, Box 0446, San Francisco, CA 94143-0446, Fax: 415-502-1411, arnold@francisco.compchem.ucsf.edu, (2) Graduate Group in Biophysics, University of California

One of the challenges facing the contemporary life and physical sciences is integrating genomic information with drug design to facilitate the development of novel therapeutic agents. Two issues that arise in addressing that goal are organizing biological structures by their binding characteristics and using that information to design and screen combinatorial libraries. We have developed a new computer program that characterizes protein structures by the shape, electrostatics, and pocket orientations of their binding sites. We will discuss using this information to form "gatekeeper" descriptions for families of binding sites that impart a hierarchical organization to protein structures and which can be used to DOCK libraries and databases against a reduced number of biological targets. We will also show how knowledge of the relative orientations of pockets in a binding site allows selection of scaffolds with geometrically complementary substitution positions. Design of scaffolds that may be general to a class of protein binding sites will be shown through a scaffold and library designed to inhibit complexes of nuclear receptors with coactivator proteins.

WEDNESDAY AFTERNOON

Section A

Patent Information Sources, Old and New

Grand Hyatt -- Constitution Ballroom A
Cosponsored with Division of Chemistry and the Law
Cosponsored with Patent Information Users Group (PIUG)
E. Simmons, Organizer

1:20 — Introductory Remarks.

1:30 — 65. Comparison of Marpat and Merged Markush Service. Andrew H. Berks, Merck & Co, 126 E. Lincoln Ave RY60-35, Rahway, NJ 07065-0900, Fax: 732-594-5832, andrew_berks@merck.com

Slides (htm)
There are two topological search engines commercially available for searching Markush structures in patents, Marpat on STN and the Merged Markush Service on Questel. This talk will compare and contrast these two search engines and their associated databases.

2:00 — 66. Using technical databases with minority patent coverage to enhance retrieval. Stephen R. Adams, Magister Ltd, Crown House, 231 Kings Road, Reading RG1 4LS, United Kingdom, Fax: (+44) 118 929 9516, stevea@magister.co.uk

Many well-established databases in science and technology include a small number of patent documents in their coverage. The proportion can vary from less than 5% to more than 20%, in subjects as varied as food science, polymer technology, energy or biology. The patent records in these databases represent an often-ignored searching source. This paper reviews the extent of patent inclusion in a wide range of technical databases, and offers some suggestions on how these records can be used in relation to the large all-patent sources such as World Patent Index and INPADOC.

2:30 — 67. Paterra^TM instant MT machine translations of Japanese patents. Alan K. Engel, ISTA, Inc, 551 W. Lancaster Ave, Suite 212, Haverford, PA 19041, Fax: 610-527-4538

Paterra^TM Instant MT machine translations is a new service for providing researchers with English translations of Japanese patents quickly and inexpensively. Researcher and information specialists access the Paterra^TM service over the Internet with standard Browsers.

3:00 — 68. New Possibilities for Linking across Patent Documents using XML and XLink. Josef M. Mattes, and Andreas Barth, FIZ Karlsruhe, Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen, Germany, jma@fiz-karlsruhe.de

Patent publications may contain two types of references to other patents: related patents (continuations and divisions) pointing to their origins and patent citations. While it is easy to follow the link in this direction the reverse direction from the original patent to the citing patent, or to the continuation, or to the division, which is of equal interest, is more difficult to perform since the links are not included in the publications themselves. Databases like Derwent's Patents Citation Index can help to find this information. In the age of the World Wide Web users are expecting to find references as hyperlinks with immediate access. Modern technologies - like XML and XML Linking Language - are the means to fulfill these expectations. This paper provides an overview of the implicit link structure of patent information and shows the various possibilities to link across the different patent documents.

3:30 — 69. Impact of patents in the CAS databases. Kerry G. Stanley, Tommy Ebe, and Jan Williams, Chemical Abstracts Service, 2540 Olentangy River Rd, Columbus, OH 43202-1505, Fax: 215-412-9375, kstanley@cas.org

CAS has covered patent information since the first issue of CA in 1907. With the increasing importance of patents as a source of chemistry-related research information and their value in supporting marketing, licensing, litigation and general business decisions, CAS has continually enhanced the completeness of its patent coverage. This session will look at the impact of the increasing volume of patent literature being processed by CAS, including the effect of patent literature versus journal literature on new CAS Registry Number entries and biosequence entries. Also discussed will be the crucial issue of currency of patent literature, approaches CAS is taking to enhance currency and the results of these approaches.

4:00 — 70. Chemical patent information from Derwent: But not directly from DWPI. John D. Myers Sr., Online Support, Derwent North America, 1725 Duke Street, Suite 250, Alexandria, VA 22314, Fax: 804-589-1895, jmyers@derwentus.com

Valuable chemical information is indexed in the Derwent World Patents Index, including biotechnology. In many cases, either subscriber indexing access or special skills are required. We will show how to use other Derwent Files to obtain DWPI originated chemical patent information. The talk will include the use of Biotechnology Abstracts, GeneSeq, the Merged Markush System and the Derwent Chemistry Resource.

4:30 — 71. Retrieving gene sequence information from patent publications. Rainer Stuike-Prill, and Anett Petersen, FIZ Karlsruhe, Hermann-von-Helmholtz-Platz 1, Eggenstein-Leopoldshafen 76344, Germany, Fax: +49 7247 808 131, rsp@fiz-karlsruhe.de

The rapid developments and the booming commercial interest in sequencing of proteins and DNA resulted in raising numbers of patent applications during recent years. As a consequence, retrieving comprehensive sets of sequence information has become a must. In addition to publicly accessible databases, i.e. Swiss-Prot, Genbank, EMBL, the GENESEQ database of Derwent Information constitutes a unique data source for biosequence information from patent publications. Available as in-house application as well as via the online host STN International, the database contains sequence data of patents from 40 patent offices starting as early as 1981. About 50% of the data are unique to this information source. Unique content and search features available in the STN implementation (DGENE) will be presented including search examples. Besides searching for bibliographic data including abstracts written by subject specialists, the most useful features facilitate searching of protein and nucleic acid sequences as well as similarity searches.

Section B

1:00 — 72. Electronic Records and Notebooks Used in R&D - Key Requirements and Initiatives. Rich Lysakowski, Director, Global Industry-Interagency Group on Electronic Records, 800 West Cummings Park, Suite 5400, Woburn, MA 01801, Fax: 781-935-3113, rich@censa.org

Collaborative Electronic Notebook Systems (CENS) are sophisticated systems for technical knowledge management that integrate electronic recordkeeping, LIMS, groupware, document management, the Web, instruments, and many applications scientists and engineers routinely use. These software systems also take advantage of handheld, wireless computer hardware. When designed and applied well, CENS greatly enhance global productivity and collaboration on complex R&D projects, while being legally reliable and trustworthy for recordkeeping. When designed or applied poorly, CENS put intellectual property, patents, and regulatory compliance -- for companies with even the best intentions -- at serious risk. This can cause major liabilities and expensive lawsuits. Developing CENS requires detailed and accurate understanding of all legal, regulatory, scientific, technology, and business requirements. This presentation will provide:

1) an overview of electronic records systems for R&D;

2) a discussion of the legal, regulatory, technical, and business requirements to implement successful systems;

1:30 — 73. Confronting the Rosetta Problem: Long-Term Access to Authentic Electronic Records. Ken Thibodeau, Director, Electronic Records Archives Program, US National Archives and Records Administration, 8601 Adelphi Road, College Park, MD 20740-6001, Fax: 301-713-6911

The National Archives and Records Administration (NARA) is the repository for the historically valuable records of all three Branches of the Federal Government. NARA's responsibilities range from guiding federal agencies in the management of current email to delivering to generations yet unborn not only rich stores of information about the national experience, but also reliable records that will stand up to judicial scrutiny. Rapid obsolescence and continuing changes in information technology pose formidable challenges to accomplishing this mission. After 30 years in the business of preserving electronic records, NARA perceives that information technology is at last on the threshold of offering viable solutions for permanent preservation and access to authentic electronic records. This presentation will describe NARA's multifaceted initiatives to build the Electronic Records Archives.

2:00 — 74. Electronic Record-Keeping: The Challenges of Imposing Standards by Regulation. David Schwarz, Office of Regulatory Management and Information, US Environmental Protection Agency, 401 M. St SW, Mail Code 21367, Washington, DC 20460, Fax: 202-260-9322

The EPA requires regulated companies to maintain records, including laboratory records, that document compliance with environmental standards. Companies must maintain these records over a period of years, both to be available for environmental inspectors, and -- at least potentially -- to provide evidence for enforcement proceedings. EPA is now writing a regulation that will set standards for these records where companies wish to maintain them electronically. This presentation will discuss the challenges of setting such standards, and especially the difficulty of reconciling the need to assure record integrity and authenticity with the diversity of business processes and technical capabilities in our universe of regulated companies.

2:30 — 75. FIRM on Electronic Records Management. Carol Brock, Overseas Private Investment Corporation, Federal Information and Records Management Council Chairperson, 1100 New York Avenue, N.W, Washington, DC 20527, Fax: 202-218-0304

A review of the Federal legislation on information management provides a broad framework for electronic records. From this review, we are optimistic about our readiness to move into e-commerce and an electronic society. Careful examination of the legislation reveals gaps in the information management laws. Reviews of state-of-the-art products and technologies show the tools must still evolve to fully satisfy Federal Requirements. These caveats are not excuses for not moving forward with eRecords; they are causes for objective planning. We can do e-business; we just need to use e-common sense. Our first e-model will not be our final model. The requirement to document the evolution of our models should be enough to satisfy the legal requirements placed on an e-society. The professional members of the Federal Information and Records Management (FIRM) Council have joined forces to enable the success of Federal ERK projects. This paper will review the FIRM's efforts to date, lessons learned, hurdles cleared, and results available.

3:00 — 76. US Patent and Trademark Office's Initiatives on Electronic Records for Intellectual Property. Art Purcell, Office of Chief Information Officer, US Patent and Trademark Office, Crystal Park 3, Washington, DC 20231, Fax: (703) 308-6879

Abstract text not available.

3:30 — 77. Auditing framework to substantiate electronic recordkeeping practices. George M. Brilis, Office of the Director, US Environmental Protection Agency - Office of Research and Development, National Exposure Research Laboratory, P.O. Box 93478, Las Vegas, NV 89193, Fax: 702-798-2233

The admissibility of electronic records has generally been allowable under the "The Business Records Exception" of the Hearsay Rule. In order to take effectively utilize this exception, electronic recordkeeping must be a part of business that is conducted as a regular activity. Providing a history of auditing recordkeeping practices may provide the evidence necessary to demonstrate that such recordkeeping is a regular activity and that those activities are of reliable quality. A framework for auditing records management is offered as a guideline in the form of a checklist, which, when utilized at periodic intervals, may provide the substance needed to prove that a high-quality level of electronic recordkeeping is a part of regular business.

4:00 — 78. Roundtable Discussion Focused on Legal and Regulatory Questions and Answers. Rich Lysakowski, Director, Global Industry-Interagency Group on Electronic Records, 800 West Cummings Park, Suite 5400, Woburn, MA 01801, Fax: 781-935-3113, rich@censa.org

This will be a moderated Roundtable Discussion focused on Questions and Answers of the attendees and presenters. We will start with a focus on known legal, regulatory, and technological issues that appear to be holding back organizations from implementing fully electronic recordkeeping and archiving systems. We will then move to questions raised by the audience as a result of the presentations.

THURSDAY MORNING

8:25 — Introductory Remarks.

8:30 — 79. Panel Presentation. The U.S. High Production Volume (HPV) Chemical Voluntary Challenge Program. Randall Brinkhuis, Office of Pollution Prevention and Toxics, U.S. Environmental Protection Agency, Washington, DC 20460

EPA ChemRTK HPV Challenge Program
HPVChallenge.com: US HPV Chemical Tracking System (industry side)
Organization for Economic Cooperation and Development (OECD) Environmental Health and Safety
Alliance for Chemical Awareness
International Council of Chemical Associations (ICCA) Chemical Tracking System
The HPV Challenge Program was created because of a recognition that basic hazard data were not publicly available for approximately 2,800 U.S. HPV chemicals (defined as those chemicals manufactured or imported at over one million pounds per year). This program is unique because: 1) it is based on a cooperative effort among government, industry, and environmental groups; 2) chemical manufacturers are volunteering to make hazard data publicly available; and 3) the data will be placed on the Internet for right-to-know purposes. It is based on the Organisation for Economic Cooperation and Development's (OECD) Screening Information Data Set (SIDS) testing program. The program includes two phases, a voluntary commitment phase, during which more than 430 companies volunteered to make data available for over 2,000 chemicals, and a test rule development phase for chemicals not sponsored by companies. [Each panel member will give a presentation on their views of the history of the program and how the information will be handled and interpreted. The views expressed are those of the authors and do not necessarily represent the organizations with which they are affiliated.]

9:30 — 80. Consumer Labeling Initiative. J. Alter¹, A. Breedlove², and M. Dominiak¹. (1) Office of Pollution Prevention and Toxics, U.S. Environmental Protection Agency, Washington, DC 20460, (2) Office of Pesticide Programs, U.S. EPA

Read the Label First! (htm)
EPA began the Consumer Labeling Initiative (CLI) in March, 1996. This is a voluntary, cooperative partnership with industry, other government agencies, and public interest groups to make health, safety, and environmental information on labels for household chemical product such as pesticides and cleaners easier for consumers to find, read, understand, and use. Extensive, peer-reviewed research with consumers was conducted to better understand current label uses and needs . These data produced a baseline against which to measure future success. Companies with over 90% of the market share in the pilot product categories actively participated. Two major reports were published - CLI Phase I in 1996, and, CLI Phase II in 1999. Recommendations for label changes are being implemented voluntarily by companies; EPA regulatory programs are changing to match.

EPA, with its Consumer Labeling Initiative partners, recently launched a major nationwide consumer education and label awareness effort to encourage consumers to "Read the Label FIRST!" The planning of that campaign, as well as its initial results, are described.

10:00 — 81. Priority-based Assessment of Food Additives (PAFA): a toxicological database. Maribeth LaVecchia, Division of Petition Control, HFS-215, U. S. Food and Drug Administration, Center for Food Safety and Applied Nutrition, 200 C Street, S.W, Washington, DC 20204

The Priority-based Assessment of Food Additives (PAFA) is a database maintained by the U.S. Food and Drug Administration (FDA) Center for Food Safety and Applied Nutrition. PAFA contains administrative, chemical and toxicological information on over 2,000 substances directly added to food, including those regulated by FDA as direct, "secondary" direct, and color additives, as well as Generally Recognized As Safe (GRAS) and prior-sanctioned substances. The database also contains less complete administrative and chemical information on 4,000 additional substances, including more than 3,000 "indirect" additives, i.e., food contact substances or constituents of such substances expected to become components of food. Limited fields from PAFA can be accessed on the Internet from the CFSAN home page at http://www.cfsan.fda.gov.

10:30 — 82. Chemical and environmental health information at the National Library of Medicine. G.F. Hazard Jr., V.W. Hudson, and P.M. Liwanag, Specialized Information Services, U.S. National Library of Medicine (NLM), 8600 Rockville Pike, Bethesda, MD 20894

11:00 — 83. Panel Presentation. Chemical accident risks: balancing the public's right-to-know against misuse of accidental release scenario data. C. Matthiessen, Office of Solid Waste and Emergency Response, Chemical Emergency Preparedness and Prevention Office, U.S. Environmental Protection Agency, Washington, DC 20460

David Wiley (US EPA, Chemical Emergency Preparedness and Prevention Office (ppt)
Chemical Emergency Preparedness and Prevention Office web site
Right to Know Network web site
OCA presentation (ppt)
OCA Fact Sheet (pdf)
The Federal Register notice that was the topic of the panel -- "Accidental Release Prevention Requirements; Risk Management Programs Under the Clean Air Act Section 112(r)(7); Distribution of Off-Site Consequence Analysis Information; Final Rule" (65 FR 48107) -- can be found at: http://www.epa.gov/fedrgstr/EPA-AIR/2000/August/Day-04/a19785.htm
or in PDF format at:
http://frwebgate.access.gpo.gov/cgi-bin/getdoc.cgi?dbname=2000_register&docid=fr 04au00-18.pdf

Representatives from the U.S. Environmental Protection Agency (EPA), Department of Justice (DOJ), chemical industry, and public interests will discuss the impact of recent regulations on public disclosure of hazardous chemical worst-case accidental release scenario assessments. Facilities submitted off-site consequence analyses (OCA) for accidental release scenarios to EPA in June 1999 as part of their required Risk Management Plans (RMPs). Although the RMPs were to be made available to the public on the Internet, concerns were raised about possible misuse of OCA information to deliberately trigger chemical releases. EPA and DOJ recently published regulations governing the distribution of OCA information in a manner that minimizes the likelihood of harm to public health and welfare from chemical releases. These regulations, as well as the risks and benefits of public disclosure of OCA data, were required by the recent Chemical Safety Information, Site Security and Fuels Regulatory Relief Act.

THURSDAY AFTERNOON

Section A

1:00 — 84. DOT's Office of Hazardous Materials Safety website. J. Albright, Research and Special Programs Administration, U.S. Department of Transportation, Washington, DC 20594

The U.S. Department of Transportation's (DOT) Research and Special Programs Administration's Office of Hazardous Materials Safety's (OHMS) Internet home page (http://hazmat.dot.gov) was established on December 1, 1996. The OHMS website is designed to disseminate information about the agency's programs and activities and to assist in complying with the Hazardous Materials Regulations (HMR). The website is updated monthly to reflect current OHMS activities and includes: Code of Federal Regulations, Title 49, Parts 100-185; recent rulemakings; exemptions and approvals; letters of clarification; upcoming public meetings, conferences and training; international activities; frequently requested documents and files; the Emergency Response Guidebook (ERG2000); incident data and Form DOT F 5800.1 and instructions; and Registration Program information.

1:30 — 85. Importance of spatial accuracy for chemical information management. G.M. Brilis, Office of Research and Development, U.S. Environmental Protection Agency, National Exposure Research Laboratory, P.O. Box 93478, Las Vegas, NV 89193, brilis.george@epa.gov

Information about chemicals can be critical to making timely decisions. The results of these decisions may not be realized for many years. In order to increase the value of information about chemicals in the environment and to create and utilize meaningful environmental models, the Environmental Protection Agency (EPA) developed and implemented a Locational Data Policy (LDP). The intent of this policy is to extend environmental analyses and allow chemical and other data to be integrated based upon location, thereby promoting the enhanced use of EPA's extensive data resources for cross-media environmental analyses and management decisions.

The EPA's new initiative, the Geographic Information Systems - Quality Assurance Team (GIS-QA), is committed to working with all organizations to ensure that spatially related tools, such as the LDP, are supported for use in gathering chemical data. An overview of the EPA GIS-QA Team and primary components of the Locational Data Policy will be presented. Internet sites will be provided for reference.

2:00 — 86. Beyond pounds: RSEI - a new tool that incorporates toxicity and exposure in screening-level estimates of risk. Nicolaas W. Bouwes Sr.¹, Richard E. Engler¹, Steven M. Hassur¹, Timothy Lehman¹, Elizabeth J. Fechner-Levy², Brad Firlie², and Susan Keane². (1) Office of Pollution Prevention and Toxics, US Environmental Protection Agency, Arial Rios Building, 1200 Pennsylvania Ave. NW (7406), Washington, DC 20460, Fax: 202-260-0816, bouwes.nick@epa.gov, engler.richard@epa.gov, (2) Abt Associates

U.S. manufacturing facilities are required to report quantities of Toxics Release Inventory (TRI) chemicals released to the environment, but these data alone do not provide information regarding risks associated with these releases. EPA's Office of Pollution Prevention and Toxics has created the Risk-Screening Environmental Indicators (RSEI) model to provide a risk-related perspective of the impact on chronic human health. EPA released this tool in July 1999. RSEI integrates toxicity scores, exposure estimates, and the exposed population to calculate risk-related scores that can be used to rank the relative impacts by chemical, geographic area, industry sector or a combination of these and other variables. The model provides users with the opportunity to examine trends, to rank and prioritize chemicals for strategic planning, to conduct risk-related targeting and to support community-based environmental protection projects. The model also allows users to investigate the relative influence of toxicity, exposure and population on the results.

Section B

2:30 — 87. 3D structural database of small molecules from traditional Chinese medicine and its use in drug discovery. Shaomeng Wang, Lei Shao, Sritha Sakamuri, and Zengjian Hu, Drug Discovery Program, Georgetown University Medical Center, 3970 Reservoir Road, Washington, DC 20007, Fax: 202-687-4032, wangs@giccs.georgetown.edu, huzj@giccs.georgetown.edu

There is an enormous interest now in the scientific community to make use of the knowledge accumulated from the Chinese and other traditional medicines for modern drug discovery. A searchable three-dimensional (3D) structural database containing approximately 10,000 individual compounds from traditional Chinese medicines has been developed. This database is expected to be useful as a link between traditional Chinese medicines and structure-based drug discovery. Examples of the use of this 3D structural database for drug discovery will be presented.

3:00 — 88. Identification, orientation, and naming of SRUs. Edward Wilks, and Manxue Wang, CRD, DuPont, BMP14, Wilmington, DE 19880, edward.s.wilks@usa.dupont.com, manxue.wang@usa.dupont.com

In the literature, we have found a disturbing number of incorrect structure-based representations according to the guidelines set forth by Chemical Abstracts Service (CAS) and the International Union of Pure and Applied Chemistry (IUPAC). This can cause problems in retrieval of polymer structures during searches.

CAS and IUPAC have published rules and recommendations for the identification, orientation, and naming of SRUs in the Index Guide and in "Nomenclature of Regular Single-Strand Organic Polymers" respectively. Although CAS names a polymer as poly(SRU) and IUPAC uses "poly(constitutional repeating unit)" or poly(CRU), the two terms are virtually synonymous. In this talk, SRU will be used. The CAS and IUPAC naming principles are essentially identical, but names of SRUs derived by the two systems are sometimes different.

In this presentation, we will review the process of how to draw and name correctly the structural-repeating unit (SRU) of a structure-based representation of a polymer and provide guidance for some commonly encountered linear polymers. Examples will be given to illustrate the three-step process.