#219 Abstracts

ACS Chemical Information Division (CINF)
Spring, 2000 ACS National Meeting
San Fransisco, CA (March 26-30)



Experimental, computational, and informatics challenges of ADME/Tox in early stage drug discovery
R. Snyder, Organizer, Presiding
9:05 1 Combining ADMET in silico, in vitro, and in vivo for drug discovery.
P. D. J. Grootenhuis, J. Penzotti, J. Miller, R. Xu, D. Kassel Molecular Design Group, CombiChem, Inc, 9050 Camino Santa Fe, San Diego, CA 92121, fax: 858-530-9998, pgrootenhuis@combichem.com
The increasing use of eADMET profiling in drug discovery requires new approaches for coupling in silico, in vitro, and in vivo methods. Ultimately, one would like to use computational models to extrapolate directly to in vivo results in humans. Because many complex mechanisms are involved in ADMET in vivo, there are multiple hypotheses that may explain the pharmacokinetic outcome. We have developed an in silico approach that analyzes a large number of hypotheses to select the set or "ensemble" of hypotheses, that, when combined, is most able to predict in vivo results for a given data set. New experimental data is collected to provide information to test our ensemble hypotheses and to iteratively refine the computational models.
9:35 2 Quantitative structure and physicochemical property-based scoring scheme to evaluate druglikeness of small organic compounds.
Arup K. Ghose, V. N. Viswanadhan, and J. J. Wendoloski. Amgen, Thousand Oaks, CA 91320, fax: 805-499-7464, aghose@amgen.com
Since the advent of combinatorial chemistry, several workers are trying to develop descriptors for drug-like molecules. This is the second effort from our group. In this approach we used structural and physicochemical aspects to develop a robust, quantitative definition of a druglike molecule. Here we used ALOGP atom type to represent the structural aspects. ALOGP, AMR, molecular weight, number of atoms and the number of heavy atoms were used to represent the physicochemical properties. A quadratic function as dictated by their distribution, was used for an iterative fitting study to differetiate the CMC (Comprehensive Medicinal Chemistry) and ACD (Available Chemical Directory) databases. This simplified scoring scheme identified 75% of CMC database and 25% of ACD database as the drug-like compounds. This general approach can be used to develop scoring scheme for any drug subclass as well as any molecular classification.
10:05 3 Computational strategies and methods for building drug-like libraries.
Tim Mitchell, Cambridge Discovery Chemistry, Merrifield Centre, Rosemary Lane, Cambridge, CB1 3LQ, United Kingdom, fax: +44 1223 722401, tim.mitchell@discoverychemistry.com, John Holland, Oxford Molecular, Medawar Centre, Oxford Science Park, Oxford, OX4 4GA, United Kingdom, and John Woods, Cheminformatics, Oxford Molecular, United Kingdom.
In order to rapidly optimise a screening hit into a candidate we need to start with the right type of compounds. Diversity should be balanced with "drug-like" properties, and the closer we get to a candidate compound, the more important "drug-like" properties become. We will describe strategies and computational methods that have been developed to rapidly assess these properties within large datasets and, in the case of combinatorial libraries, methodologies that rapidly identify reagents that result in products with undesirable properties such as toxicity and poor ADME profiles.
10:35 4 Design of leadlike combinatorial libraries
Simon J. Teague and Andy Davis. Charnwood, AstraZeneca R&D, Bakewell Rd, Loughborough, Leicestershire, LE11 5RH, United Kingdom, fax: 44 01509 645571,http://www.lib.uchicago.edu/cinf/219nm/%20mailto:simon.teague@astrazenec...
Why has the combinatorial chemistry-high throughput screening paradigm of drug discovery produced so few development candidates so far? The Design of lead-like combinatorial Libraries is described. These are defined as consisting of compounds with M.Wt. < 350, clgP < 3.0. If members of this library display micro molar affinity in an HTS screen then these properties allow scope for optimisation of potency and physicochemical properties during their subsequent elaboration into drugs. Acceptance of these design parameters may justify re-examination of the role of some combinatorial technologies. Large libraries generated by concatenating several monomers using multi-component reactions or using several steps incorporating split and mix protocols, may not be the most effective approach. The impact up the relative merits of solid and solution phase approaches to library synthesis, diversity measures and analytical techniques will be briefly discussed.
11:05 5 Using LeadScopeTM for exploring large sets of toxicology data
Paul E. Blower1, Kevin P. Cross2, Wayne P. Johnson1, Glenn J. Myatt1, and Gulsevin Roberts1. (1) Columbus Molecular Software, Inc, Columbus, OH 43212, fax: (614) 675-3732, pblower@columbus.rr.com, (2) Research and Development, Columbus Molecular Software, Inc, Business Technology Center, 1275 Kinnear Rd, Columbus, OH 43212
Modern approaches to drug discovery have dramatically increased the speed and quantity of compounds that are made and tested for biological activity. LeadScopeTM is decision support software to assist pharmaceutical researchers in mining knowledge from the huge volumes of data now being generated by these high-speed techniques. It performs a systematic substructural analysis of a compound set using a class hierarchy containing over 27,000 familiar structural features. Sophisticated data visualization tools assist the scientist to interactively explore structures by properties and common structural features, helping to quickly find correlations between biological activity and structural features. In this talk, we will illustrate the use of LeadScopeTM to explore a large toxicology dataset and locate features or combinations of features that are highly correlated with toxicity.
12:00   Lunch Break
1:30 6 Fast computational filters for predicting ADME/Tox
Jens Sadowski, ZHF/G - A 30, BASF AG, Ludwigshafen D-67056 Germany, fax: +49/621/6020440, mailto:sadowski@zhs4.zh.basf-ag.de

The enrichment of compound collections for synthesis by combinatorial chemistry, for purchase, or for biological testing in high-throughput screening with compounds being drug-like, bioavailable, stable, and non-toxic becomes more and more a new focus for computational chemistry. Whereas simple rules for bioavailability are published [1] and filters that can distinguish between drugs and non-drugs recently became available [2], there is no general set of ADME/Tox filters available.

We have established a general method for constructing such filters based on neural networks and large amounts of data from structural databases or HTS. In analogy to the drug/non-drug filter the method generates fast approximate computational filters that can discriminate between bioavailable/non-available compounds, toxic/non-toxic compounds, etc.

[1] Lipinski et al., Advanced Drug Delivery Reviews 1997, 23, 3.

[2] Sadowski et al., J. Med. Chem. 1998, 41, 3325.

2:00 7 Computational models for the prediction of pharmacokinetic properties at Pharmacopeia Inc
William J. Egan and Kenneth M. Merz Jr. Center for Informatics & Drug Discovery, Pharmacopeia, Inc, 2000 Cornwall Road, Monmouth Junction, NJ 08852, fax: 732-422-0156, wegan@pharmacop.com
Considerable attention has been focused recently upon the task of predicting pharmacokinetic properties of compounds early in the drug discovery process, when failure of a lead due to poor ADME properties is far less costly. We will discuss Pharmacopeia’s efforts to prioritize the development and optimization of active compounds using predictive ADME models. Results of physicochemical based QSPR models predicting intestinal absorption, blood-brain barrier penetration, and other pharmacokinetic properties will be presented for literature compounds, individual Pharmacopeia compounds, and Pharmacopeia ECLiPSTM libraries.
2:30 8 Computational modeling and prediction of human intestinal absorption and blood-brain barrier penetration.
Jie Q. Wu, Camitro Corporation, 4040 Campbell Avenue, Menlo Park, CA 94025, jie.wu@camitro.com
Purpose: To statistically model the predictive quantitative structure-activity (QSAR) relationships for (a) human passive intestinal absorption, Fa; (b) blood-brain barrier (BBB) penetration, logBB. Methods: Two structurally diverse sets of molecules were investigated. Selection of training/test sets, and theoretical computed molecular descriptors, and relationships between (a) Fa, (b) logBB and molecular descriptors, were accomplished by Multivariate Statistics (PLS, D-optimal design, Clustering and Discriminant Analysis). Results: Excellent statistical models were achieved with easily rationalized molecular descriptors. The results together with prediction of external data sets were discussed and compared with updated literature results. Conclusions: The results provide a rational basis for understanding mechanisms of drug transport across (a) intestinal barrier, (b) BBB and how to alter chemical structures to attain improved oral drug absorption and brain penetration at a much earlier stage drug discovery (e.g. in virtual library design and virtual HT screening).
3:00 9 Predictive modeling of cytochrome P450 mediated drug metabolism
Todd Ewing, Jie Wu, Jean-Pierre Kocher, and Ken Korzekwa. Computation Chemistry, Camitro Corporation, 4040 Campbell Ave, Menlo Park, CA 94025, fax: 650-327-4639, todd.ewing@camitro.com
The interaction of drug candidates with cytochrome P450 enzymes has become an important consideration in the pharmaceutical design process. The development of predictive models of P450 drug metabolism based on enzyme binding and substrate reactivity will be discussed. An approximate binding model first identifies whether a compound is likely to bind as a substrate or inhibitor to each of the most important isoforms of P450. Second, a quantum mechanical reactivity model identifies whether a compound with binding activity may be a substrate for P450 metabolism by either hydrogen abstraction or aromatic oxidation. Third, a detailed binding model for each P450 isoform identifies which reactive sites on a potential substrate would be suitably positioned for attack by the enzyme. The application of these models toward library design and lead optimization will be discussed.
3:30 10 Blood levels of compounds: Structure property relationships
Daniel John Russell, Physical Sciences, AstraZeneca, 1800 Concord Pike, P. O. Box 15437, Wilmington, DE 19850-5437, daniel.russell@phwilm.zeneca.com
The Atlas Do structural features correlate with the blood levels that are observed after oral dosing ? How do the aqueous solubility and logD of a compound influence the observed blood level? These questions can be addressed by studying the results from a High Throughput Blood Level Assay performed on a wide variety of compounds at AstraZeneca. Results and conclusions will be presented.
4:00 11 Structure-based computational models of cellular permeability
Jay T. Goodwin, Boryeu Mao, Thomas J. Vidmar, Philip S. Burton, and Robert A. Conradi. Drug Absorption and Transport, Pharmacia & Upjohn, Computer-Aided Drug Discovery, and Biostatistics, 7271-267-606, Kalamazoo, MI 49007, fax: (616) 833-2325, jay.t.goodwin@am.pnu.com
Predictive models of cellular and intestinal permeability are of potentially significant use in drug discovery and design. Through the design and use of model peptides we have defined several solute physicochemical determinants of permeability, including lipophilicity and hydrogen-bond desolvation potential, in the Caco-2 cell monolayer system. Based upon these results we have begun to evaluate structure-based computational models of permeability using both simpler geometric and more complex, solvation-based descriptors. Preliminary results from these modeling studies will be compared and contrasted.




Chemical information in the 21st Century
R. Snyder, Organizer, Presiding


Introductory Remarks


Moving toward a new digital environment
Robert J. Massie, Chemical Abstracts Service, P.O. Box 3012, Columbus, OH 43210, fax: 614-447-3765, rmassie@cas.org

In the coming years, vast and extraordinarily varied sci-tech data repositories will become available on the Web and its descendants. This paper surveys new attempts by many industry participants to create what has until recently been only a dream - truly integrated digital sci-tech research environments. These integrated environments may take on increasingly global and pan-scientific characteristics as the cost of linking and data storage continue to decline. All the major publishing conglomerates are pursuing this objective. Surprisingly, so are certain agencies of the US Government, which believe that their missions include providing taxpayer subsidized scientific information on the Web.

In the past, the emphasis for information services was on "value-access". In other words, providing high quality access to "difficult to find or retrieve" information was the central benefit. A new perspective on the future is explored wherein the emphasis for information services will be "value-aggregation". In other words, how well does the service bring together information that is otherwise available on the Web or elsewhere? How much more convenient, valuable and productivity-enhancing is the totality of the service provided, even if individual information elements are "free" elsewhere on the Web? The most effective and earliest "value-aggregators" will establish strong installed bases in the new Web world and have the opportunity to grow with their customers and take advantage of the continuing information industry evolution. CAS initiatives as one of these value-aggregators are discussed in the context of other developments taking shape in the new era of chemical information.



Journal publishing at the ACS in the second millennium
Ralph E Youngen1, Lorrin R. Garson1, and Susan Barclay2. (1) Information Technology/Publications, American Chemical Society, 1155 Sixteenth St., NW, Washington, DC 20036, r_youngen@acs.org, (2) New Product Development, American Chemical Society
It is clear that journal publishing is undergoing a remarkable transformation as we enter the second millennium. Electronic delivery is rapidly gaining acceptance yet a myriad of challenges faces the scientific community. Among these are a variety of business and technical issues: What is the electronic journal likely to metamorphose into? What kinds of scientific data are most amendable to electronic dissemination and in what formats? How will publishers transform data through succeeding generations of file formats and delivery systems? What is an acceptable format for the digital archive and who should have access under what terms? Through what business model will publishers generate revenues and what are users willing and able to pay. An overview of industry-wide electronic publishing practices will be presented with prognostications for the first decade of the second millennium.
9:05 14 Managing and utilizing chemical/biological data in the age of combinatorial chemistry/high-throughput screening
John J. Baldwin, Pharmacopeia Inc, CN 5350, Princeton, NJ 08543-5350, fax: 609-452-3672, baldwin@pharmacop.com
Chemists and biologists are accustomed to managing drug discovery projects on spreadsheets that allow the manipulation of hundreds of compounds over a limited range of bioassays. Today with combinatorial collections counting up to several million compounds and potentially hundreds to thousands of assays, informatics specialists are faced with extraordinarily difficult cheminformatics challenges.

From the users point of view, such informatics systems must support registration/tracking of an enormous, ever-growing number of compounds, structural confirmation data, and results from screening assays/follow-up studies. Library design must be aided by rapid computer facilitated analyses which support: diverse synthon selection, drug-like/ADME characteristics from single compounds to virtual libraries and diversity/similarity comparisons across compound collections. Such systems ideally will handle data warehousing with rapid data mining capabilities and auto-pharmacophor generation/comparison to aid in the selection of subsets having higher success probability against novel targets.

As these tools evolve, the medicinal chemists will have on their desktops all the assets needed to design bioactive compounds and libraries which have a higher probability of avoiding the downstream issues of bioavailability and toxicity.

9:35 15 Evolution from journal article to knowledge base
Patrick Jackson, Elsevier Science B.V, Molenwerf 1, 1014 AG Amsterdam Netherlands, fax: 31 10 485 2845, p.jackson@elsevier.nl
Readers and users are increasingly demanding electronic access to chemistry full text literature, which is accompanying conventional access through printed media. The intrinsic value of such documents, particularly for those working in multidisciplinary fields, is greatly increased when the user also has access to a very large universe of electronic documents in related areas, preferably on the same platform under one licensing arrangement, as is offered, for example, by accessing the ScienceDirect database. A richer and deeper result is obtained by reference linking both to related full text documents as well as to databases providing supplementary information, utilising both static and dynamic linking methods. Readers' expectations are thus changing away from the concept of single-article, single-journal access towards multi-article, multi-journal, multi-information type access. Publishers will need to follow a policy of increasing integration of their primary, secondary and even tertiary information sources (such as Major Reference Works) to be able to satisfy reader demands fully. A variety of new business models will develop over time to provide greater flexibility for the purchaser.


Information management as a platform supporting pharmaceutical discovery.
John P. McAlister, Tripos, Inc, 1699 South Hanley Road, St. Louis, MO 63144, fax: (314) 647-8108, johnmc@tripos.com
Discovery of new products in the pharmaceutical and related industries has traditionally been viewed as a purely scientific endeavor with independent creative contributions from a number of disciplines. Success has been usually been described as "serendipitous." Forward-looking organizations have refocused their thinking on discovery, viewing it as a business process that can be understood and optimized. Key to the success of this effort is an informatics platform spanning the entire discovery process from target validation through pharcokinetics and toxicity studies. This requires a new paradigm for data management and analysis. An approach to this challenge that is achieving increasing acceptance will be described as well as its implications for the future of drug development.
10:35 17 Mining the discovery data mountain
David Jackson, Oxford Molecular Group PLC, Medawar Centre, Oxford Science Park, Oxford OX4 4GA United Kingdom, fax: +44 1865 784601, djackson@oxmol.co.uk
The last ten years has witnessed a revolution in the experimental life sciences, offering unprecedented opportunities in discovery research. The transition to automated biological and chemical sciences is unleashing data, both public domain and proprietary, structured and unstructured in composition, in exponentially increasing volumes. The effective capture, management and mining of this data mountain represents one of the most significant challenges to the pharmaceutical industry in its drive for efficiency gains in the discovery process. This talk will highlight some of the issues involved and demonstrate new ways of data management and the implications on the R&D process.
11:05 18 Discovery informatics framework: Delivering the integration promise
Patricia Rougeau, MDL Information Systems, 14600 Catalina Street, San Leandro, CA 94577, prougeau@mdli.com
Organizations that can turn all of the information assets available to them into actionable decisions will have the competitive advantage in the future. The web, internal corporate databases, reports, information previously unused or isolated in context, all contribute to the overwhelming amount of information which can be assimilated. The key is to integrate public and private information smoothly into the workflow of the scientist.

An effective informatics infrastructure is not an option, but is as critical to an organization's success as the underlying enabling technologies such as automation, miniaturization, and robotics techniques applied to chemistry and screening. An informatics infrastructure can organize and enhance the ad hoc scientific process, providing process discipline to support creative science.

Ultimately, the discovery informatics framework links business concerns and scientific concerns in ways never before possible. Organizations who achieve this link will realize real economic benefits in the 21st century.

11:35 19 Panel Discussion
R.W. Snyder
A panel discussion will be held with all speakers of the session acting as panel members. Members of the audience may ask questions to one or more of the panel members.



Section A


Experimental, computational, and informatics challenges of ADME/Tox in early stage drug discovery
3. Plenary Session
O. F. Güner, Organizer, Presiding


Introductory Remarks
1:05 20 Changes in the profiles of drug properties: An experimental, computational, and informatics perspective.
Christopher A. Lipinski, Pfizer Central Research, Eastern Point Road, Groton, CT 06340, fax: 860-715-3345, lipinski@pfizer.com
Drug properties have changed so that discovering orally active drugs is more difficult. Compared to phase II drugs, aqueous solubility is lower in commercial drugs, in HTS leads, and in compounds from HTS followup. Chemistry isolation by crystallization biases towards poor aqueous solubility while tight SAR feedback from in-vivo oral testing biases towards good aqueous solubility. Lead identification strategy affects the relative importance of poor permeability or poor solubility. Leads from structure based drug design tend towards larger size, increased hydrogen bonding potential and poor permeability. HTS based leads tend towards larger size, higher lipophilicity and poor solubility. Computational filters such as the "rule of 5" flag compounds with profiles that are most problematic for oral activity. Large stratified drug databases provide an informatic perspective on important non physico-chemical, biologically based, attributes of drug quality and survivability.
1:50 21 Bioavailability consideration in drug design
John J. Baldwin, Pharmacopeia Inc, CN 5350, Princeton, NJ 08543-5350, fax: 609-452-3672, baldwin@pharmacop.com
The earliest stages of the drug discovery process, i.e., target identification/validation and lead discovery/optimization, in large part, dictate a project’s ultimate success or failure. Drug discovery is a lead driven process that is moving from its purely potency focus to one that incorporates ADME considerations and involves a greater emphasis on early interdisciplinary considerations. The driving force for this change in emphasis has been an economic one, that is, an attempt to eliminate failures at the early, least expensive phase of the discovery process. The approach to achieve this objective starts with improved library and compound design guided by high throughput bioanalytical and computational methods which can rapidly define parameters believed to be important to bioavailability. Continued advances in rapid predictive models to serve as guides to the medicinal chemists are critical to accelerating the discovery process and increasing the probability of success
2:25 22 e-DDI: The virtual discovery-development interface
Dale E. Johnson, Chiron Corporation, 4560 Horton Street, Emeryville, CA 94608-2916, fax: 510-923-7809, dale_johnson@cc.chiron.com
The discovery development interface (DDI), which spans hit-to-lead optimization to early clinical trials is characterized by a high drop-out rate and less than optimal predictions for later stage clinical outcomes. Key DDI questions continue to remain unchanged: What compounds should be made next? Will the next compound have good "drug-like" properties? Will predicted toxicities be acceptable for development? And will there be an acceptable therapeutic index in specific patient populations? However, the means to answer these questions are changing rapidly due to emerging in silico technologies where virtual chemical libraries can be screened in virtual tests and simulation algorithms can answer key DDI questions--prior to making the next compound. Transitioning DDI into e-DDI requires a clear definition of the e-DDI toolbox, and most importantly, simulation algorithms must be based on quality/diverse data where models are tested with chemical libraries designed to answer chemical space and biological response questions.
3:00 23 Changing the paradigm to reduce drug candidate attrition: A toxicology perspective
Stephen K. Durham, Bristol-Myers Squibb PRI, P.O. Box 4000, Mailstop F14-03, Princeton, NJ 08543, fax: 609-252-7156, stephen.durham@bms.com
Changes in synthetic chemistry strategies, coupled with enormous developmental costs, has forced dramatic changes in the drug selection paradigm. This paradigm shift necessitates the early and successful selection of drug candidates, and the identification of toxicologic liabilities to eliminate failures during the initial and least expensive phase of the discovery and development process. The advent and utilization of commercially available, predictive toxicology computational programs is taking enormous steps towards filling this void. However, many of these programs, as they relate to the pharmaceutical industry, are poorly validated and use inappropriate data sets. Additional avenues of pursuit include the design of proprietary libraries that are enriched with "institutional knowledge". In the interim and until robust, validated in silico methods are routinely incorporated in the evaluation process, predictive in vitro assays and emerging molecular technologies must be aggressively utilized in the evaluation process.
3:35 24 Role of informatics and computational tools in optimizing ADME/Tox properties
Mark A. Murcko, Vertex Pharmaceuticals, Cambridge, MA 02139, fax: 617-577-6680, murcko@vpharm.com
Recent work from several labs suggests that computational methods, such as expert systems, may be able to make reasonable predictions on the physical and pharmacokinetic properties of drug candidates. This capability holds the potential to have a huge impact on the overall efficiency of drug discovery. We will describe some of these early efforts, point out the current limitations of "in silico" predictions, and suggest future directions for the field. In particular, the importance of having access to much larger databases of experimental data with which to parameterize and validate the computational methods will be discussed.




Sci-Mix Poster Session
A. H. Berks, Organizer, Presiding
7:00 PM - 9:00 PM
  25 Beyond the molecular structural reaction search
Helen Weiqin Yun, Wyeth-Ayerst Research, 401 North Middletown Road, Pearl River, NY 10965, fax: 914-732-3744, yunh@war.wyeth.com
Many databases for reaction searching are currently available, including CASREACT, CHEMREACT, REACCS (the combination of several smaller databases such as Organic Syntheses, Theilheimer, JSM from MDL), CHEMINFORMRX, and Beilstein. Because of the differences in time coverage, indexing policies, etc., it is necessary to search all available databases for comprehensiveness. In addition, there are default definition differences in these databases in structure representations, query construction, and searching mechanisms. Some articles have compared differences between reaction databases and advantages of using particular databases for particular purposes. This poster uses an example searching CASREACT, and Chemical Abstracts to illustrate the effects of database defaults and provides suggestions on avoiding pitfalls caused by these defaults.
  26 Data transfer routine for spreadsheet software
Perry M. Koussiafes, Fire and Arson Laboratory, Florida State Fire Marshal, Rt.1 Box 3252, Hwy. 90 West, Havana, FL 32333, fax: 850-539-9662, koussiafesm@doi.state.fl.us
Spreadsheet software is commonly used to perform calculations and generate reports on data collected from instruments in the laboratory. The data is often typed into the spreadsheets manually, a labor intensive process that is susceptible to data entry errors ("typos"). Data collection software usually allows the option of saving the data in an ASCII text type file. A simple routine to transfer data directly from ASCII text files to the spreadsheet can eliminate many of the data entry errors as well as improve the efficiency of the data entry process.

This poster will present a simple data transfer routine that can be incorporated into a spreadsheet. With some modification it may be adapted to other software products as well. I will review the requisites for data transfer, present the source code, and demonstrate how to implement this source code.

  27 Silicon oxidation reaction details: Searching for resources
Stella Ota, Stanford University, MC 2125, Bldg 380, 4th Floor, Stanford, CA 94305-2125, fax: 650-725-8998, sota@sulmail.stanford.edu
Oxidation of semiconductors occurs without fanfare in clean rooms worldwide. However, retrieving reaction details for the oxidation of an inorganic substance such as silicon proves illusive. With the exception of Gmelin, many resources for searching chemical reactions exclude inorganic and organometallic reactions. Unfortunately, even in those resources covering non-organic reactions, a detailed reaction component such as Si{100} cannot be specified. This poster covers search strategies and resources for silicon oxidation reaction details.



Section A


Chemical Information Needs at the Graduate Level
G. Baysinger, Organizer, Presiding
8:25   Introductory Remarks
8:30 28 Chemical information course for graduate students in organic chemistry.
Kierstin Child, Chemical Sciences Subject Specialist, Brigham Young University, 2520 Harold B. Lee Library, Provo, UT 84602, fax: 801-378-6708, kierstin_child@byu.edu
Prior to 1999, there was no established library instruction program for the graduate students in chemistry at the author's institution. Graduate students in organic chemistry are required to complete either a foreign language requirement or an online searching skills requirement prior to obtaining their advanced degree. Most student choose the latter option, which has consisted of a brief set of written instructions and a few exercises to complete. In cooperation with a faculty member from the chemistry department, a chemical information instruction course has been created that introduces the students to significant resources in the library and teaches the skills of online searching with emphasis on techniques for searching the CA files through STN International. The course runs approximately eleven weeks and fulfills the online searching skills requirement for obtaining their degree. Response from the students has been overwhelmingly positive. HTML version of slides available.
9:00 29

Dealing with informational overload in chemistry: A graduate student view.
Roman A. Boulatov, Department of Chemistry, Stanford University, Stanford, CA 94305-5080, fax: 650-725-0259, boulatov@leland.stanford.edu
There are approximately 14,000 chemistry journals worldwide, with an average of 800,000 pages printed daily. For graduate students, who must master the fundamental theories and facts that have emerged in the past and remain current, these numbers represent a huge problem. No ideal solution exists. Appropriate databases and the skills to search them are crucial, but not a panacea. The fundamental problem with indexes is their subjectivity. Full-text search available for ACS journals is an improvement but goes back to only 1996 and covers one journal at a time. Those of us aiming at a professorship in top-tier universities face a challenge of keeping current in a range of subfields, for the most significant discoveries are often made in the boundaries between them. Research advisors, not electronic databases, are critical in facilitating this, by bringing to their students' attention important articles in the fields beyond these student projects.

9:30 30 Education of chemistry librarians and chemical information specialists in the 21st century
Gary D. Wiggins, Chemistry Library, Indiana University, 800 E. Kirkwood Avenue, Chemistry Building Room C003, Bloomington, IN 47405-7102, fax: 812-855-6611, wiggins@indiana.edu and Charles H. Davis, School of Library and Information Science, Indiana University, 1320 E. 10th Street, Main Library 005F, Bloomington, IN 47405-3907.
Traditional library science programs have changed dramatically in the last decade of the 20th century. Although Indiana University's School of Library and Information Science has modified its program to include increased emphasis on information science, a new School of Informatics will offer additional avenues of specialization for chemists who want to pursue careers in chemical informatics. The development of the school and the planned courses leading to the BS and MS degrees in chemical informatics will be described. In addition, the results of a survey of practicing chemical information specialists and librarians on the factors which led them to choose their professions will be presented.
10:00 31 Impact of Minerva/Beilstein CrossfireTM on provision of services to graduate students
Susanne J. Redalje, Chemistry Library, University of Washington, Box 351700, Seattle, WA 98195-1700, fax: 206-543-3863, curie@u.washington.edu
Many schools across the U.S., large and small, are providing access to chemical data and literature via Beilstein Crossfire TM. The Minerva Consortium, with its scalable pricing, made this possible for libraries who previously would have been unable to provide this much access. Beilstein Crossfire TM provides access to two databases, Gmelin and Beilstein, which cover literature from the late 1700s and early 1800s. Access to older literature has proven valuable to users but also has implications on the future of how libraries provide access to this information. Two areas, in particular, are impacted: storage of older materials when space is a consideration and access to older materials which are not owned by the institution. More information available.
10:30 32 Not too remote: Strategies for promoting chemical information services outside the library
Andrea B. Twiss-Brooks, John Crerar Library, University of Chicago, 5730 S. Ellis, Chicago, IL 60637, fax: 773-702-7429,mailto:atbrooks@midway.uchicago.edu
As libraries increase the amount of chemical information that they provide via web and via other remote access methods, graduate students in chemistry may have fewer opportunities for person to person encounters with library staff. In order to keep students (as well as other users) informed of available databases and online journals, to provide search assistance, and to insure efficient delivery of chemical information, it is essential to establish good lines of communication. In a primarily research institution that lacks a formalized structure for chemical information instruction, this communication must take place outside the library walls. This presentation will present an overview of efforts to increase communication with students about library resources and services.
11:00 33 Preparing students for the graduate research library
Patricia O'Neill, Mortvedt Library, Pacific Lutheran University, Tacoma, WA 98447, fax: 253-535-7315, kirkwope@plu.edu
As the availability of electronic resources in many graduate programs increase small undergraduate programs face a new version of the old dilemma. Without the research funding it is difficult to justify large expenditures for research tools which do not directly support the core curriculum. What, of the resources can we afford, should be used to prepare students for the graduate research library? With paper resources, a few could be used as examples of the range of materials available. Concepts of searching one paper resource could be translated to others. Teaching the paper tools is no longer a viable option. Students expect electronic resources. Undergraduate librarians must find innovative solutions to this problem.
11:30 34 Where'd everybody go? Reflections on the outcomes of the virtual library
David Flaxbart, Chemistry Library, University of Texas at Austin, Welch Hall 2.132, Austin, TX 78712, flaxbart@uts.cc.utexas.edu
As the virtual library becomes a reality in content as well as in concept, we are beginning to witness shifts in the way chemistry libraries are used, particularly by our most avid patrons, graduate students. Users who once lived in the library now access more of what they need remotely -- catalogs, indexes, journals, reserves, and soon possibly books. Are they visiting the library less? Are they losing or never learning library skills as they gain digital skills? Are they missing too much of the literature? Do they care? Should we care? This presentation will examine some numbers to determine if usage patterns are indeed changing, and what those changes might mean for libraries that serve chemistry students.



Section B


Joint CSA/CINF Symposium: Chemical Information and E-Commerce 1
B. A. Vickery, Organizer, Presiding
8:25   Introductory Remarks
8:30 35 Integration of technical information with on-line buying increases the chemical procurement efficiency
John Custer, Internet Strategies, Sigma-Aldrich Research, 3050 Spruce Street, St. Louis, MO 63103, jcuster@sial.com
International Data Corporation predicts that by 2003, 80 percent of all business transactions will be carried out on-line. Chemical manufacturers and distributors hoping to tap into this trend must build web sites that emulate the entire procurement process from product selection, to order placement and tracking, to product and order support. This can only be accomplished by creating an on-line buying experience that integrates Physical Properties, Product Specifications, Safety Information, Product Applications, Availability, Delivered Price, Order Tracking, and Customer and Technical Support. Our new web site release combines this eCommerce functionality with a robust search engine that allows researchers to locate products by name, structure, CAS number, molecular formula and other important criteria.
9:00 36 Integrating reagent selection, ordering, and inventory management
Charles D. Sullivan, Maurizio Bronzetti, and David A. Evans. MDL Information Systems, Inc, 14600 Catalina Street, San Leandro, CA 94577, fax: 510-614-3651, charless@mdli.com
Enhanced reagent selection, cost reduction and increased productivity can be achieved by the effective use of chemical information systems. These enable the identification of the best reagents to build chemical libraries, reduce the time and cost of locating in-house chemicals or selecting and ordering chemicals from a preferred supplier. Integration with purchasing and inventory management systems provides even greater benefits including helping to reduce costs, meet regulatory requirements and reduce waste. This presentation will discuss some of the newest integrated tools available for reagent selection, ordering and chemical inventory management.
9:30 37 Chemical information and e-commerce in the 21st century: The role of the infomediary in online commodities trading
Arlyn R Sibille, Vice President Analytical Testing, WorldWideTesting.com, 430 Tenth St. NW, Suite N-104, Atlanta, GA GA 30318, fax: 404-815-5851, arlyn.sibille@worldwidetesting.com
Online purchasing of chemicals can be a risky and time-consuming job. Suppliers waste time giving information to consumers who may not buy. Buyers spend time and money searching for a supplier that fits their needs and then verifying the integrity of the products. "The online commodities market is exploding. Hundreds of business-to-business e-marketplaces already exist," said Leah Knight, e-commerce analyst of the Dataquest unit of Gartner Group. "This drives a significant need for a nonpartisan entity that truly protects both the buyer and the seller. Any company that does this has a very promising future and will be a must-have for online commerce." A new term, 'infomediary' has been coined to describe such an impartial entity and a company has evolved that fits the description, WorldWideTesting.com.
10:00 38 Demand aggregation in real-time as a new model for e-commerce
Lane S. Yago, PowerPurchasing Group, 1840 San Miguel Drive, Suite 100, Walnut Creek, CA 94596, fax: 925-042-0632, lyago@powerpurchasing.com
Electronic Commerce (e-commerce)has emerged as a viable form of business. The leading edge of e-commerce is the "exchange" model, which consists of a bid-ask scenario. Demand Aggregation is a form of the exchange model, where individual purchases (Demand) are aggregated into a larger order and then submitted electronically to a group of vendors for bidding. This type of e-commerce is uniquely suited for small to mid-sized organizations that typically do not enjoy a great deal of savings from suppliers. Indeed, many suppliers have a annual volume threshold, below which they do not even call on smaller organizations. E-commerce in the form of Demand Aggregation changes this; as the internet has demonstrated, there is a significant shift from a seller-centric world to a buyer-centric virtual world. While the typical scenario describing the buyer-centric nature of the internet has to do with a buyer having the power to buy from anyone at anytime and anywhere, Demand Aggregation has the additional dimension of providing small business to business (B2B) buyers the ability to group together on-the-fly and use their collective purchasing power to their advantage individually as well as collectively. Web-based purchasing systems that are dynamically driven by the buying activities have been developed that submit volume purchases to an artificial intelligence-based search program that polls multiple suppliers and returns price, availability, part number match as well as other key metrics. The system is designed to operate 24/7 and is not constrained by any particular industry. Examples will be discussed where Demand Aggregation has provided a significant reduction in item prices as well as drastically reducing the cost of procurement experienced by the individual buyers.
10:30 39 Bringing clarity to e-commerce in the industry
Brad Lich, vice president of marketing, e-Chemicals Inc, 505 E. Huron St, Suite 208, Ann Arbor, MI 48104, blich@e-chemicals.com and Mike E. Efting, president, e-Chemicals Inc, 505 E. Huron St, Suite 208, Ann Arbor, MI 48104.
e-Chemicals designed an e-commerce chemical marketplace by following the principle that technology should be adapted to meet the specific needs of the chemical supply chain. Recognizing that buyers and sellers of industrial chemicals could realize cost savings through the efficiency of e-commerce transactions and more sophisticated supply chain management, e-Chemicals offers two methods of e-commerce purchasing.

e-Chemicals offers an online solution, www.e-chemicals.com, where pre-qualified purchasers choose to buy through a multi-vendor on-line catalog, at an auction, or through an exchange. The solution allows buyers to select a product, get a price, order and track shipments online. The Web site offers product information, industry news and feature reports on critical issues that impact the chemical marketplace.

Additionally, e-Chemicals offers an e-procurement system for buyers who want to integrate their chemical purchasing supply chain with an existing ERP system. Specifically, e-Chemicals works with the purchasing organizations to streamline the procurement process by eliminating as many steps as possible and automating the rest.




Joint CSA/CINF Symposium: Chemical Information and E-Commerce 2
B. A. Vickery, Organizer, Presiding


Introductory Remarks
2:15 40 ChemNavigator.com: An iResearch™ system for the acquisition of compounds for pharmaceutical lead follow-up
Tad Hurst, ChemNavigator.com, 6166 Nancy Ridge Road, San Diego, CA 92121, fax: 858-625-2377, THurst@ChemNavigator.com
This new iResearch™ system allows researchers to find and purchase a set of compounds related to a lead structure for follow-up testing. The compounds are chosen by the user from a variety of sources using similarity comparisons and analog searching. ChemNavigator maintains data of the availability of the compounds and their resupply status to ensure timely delivery (1-2 weeks) of the structures selected. In this presentation we will discuss the methods used for automatic filtering to remove non-drug-like compounds according to standard and user-customizable rules, as well as the similarity-based clustering used to display the hits from searches of the databases.
2:45 41 E-commerce models: Changing the way vertical industries do business
Kevin Wenta, CheMatch.com, 2900 North Loop West, Suite 1120, Houston, TX 77092, cwilliams@piercom.com
New online business models are reshaping vertical industries, delivering unprecedented efficiencies to the commerce process. For example, e-commerce is changing the way the $400 billion-a-year global petrochemical market buys and sells large quantities of bulk commodities. According to Forrester Research, petrochemicals is expected to have online sales of more than $150 billion in 2003, or 17% of the Internet economy. E-commerce brings tremendous value to the industry. In addition to uniting buyers and sellers worldwide, chemical executives can discover real-time pricing and market transparency to manage inventory, supply/demand imbalance, financial exposure and to optimize cash flow in a cyclical business. Thanks to the Internet's efficiency, chemical companies can realize an uplift in bottom-line operating income of 15 percent. In today's economy, companies need immediate round-the-clock access to trusted, actionable information, coupled with real-time trading capabilities. The appropriate first-mover business model has a head start in creating a truly functioning market unfettered by the boundaries of time, space and geography, and in applying the model to other vertical industries.
3:15 42 Leveraging e-procurement in pharmaceutical R&D
Mark Robillard, Vice President of Electronic Commerce, EMAX Solution Partners, 18 Campus Boulevard, Newtown Square, PA 19073, fax: 610-325-3782
The anticipated explosion in Business to Business electronic commerce has sent most pharmaceutical procurement organizations in search of software and services to leverage this burgeoning digital supply chain. Concurrently, pharmaceutical R&D operations are facing increasing pressure to increase the discovery of the new compounds through highly automated discovery factories.

This talk will explore the primary drivers behind the digital procurement movement, the emerging choices and landscape of solution providers. In addition, key considerations on how to integrate these solutions to improve research productivity will be presented.

3:45 43 Online marketplace: Strategies for reaching customers through electronic commerce
David A. Weber, Supplier Relations, Chemdex, 3950 Fabian Way, Palo Alto, CA 94303, fax: 650-813-0304
With the emergence of the Internet economy, the chemicals industry has begun to see dramatic changes in the traditional buyer-seller relationship as suppliers and buying organizations realize the unprecedented opportunities of electronic commerce. In a panel discussion, representatives from Chemdex and industry suppliers that may include Incyte Pharmaceuticals, Bayer Corporation's Ion Exchange and Water Treatment Group,Greiner America and BioWhittaker will address the following:

*The advantages of electronic commerce for chemicals suppliers--strategies for reaching a broader customer base *Offering greater breadth of customer services through electronic commerce *The online marketplace as a sales and marketing partner *Implications for the future of electronic commerce in the chemicals industry

4:15 44 Proven benefits of e-commerce for chemical buyers and sellers
John Beasley, ChemConnect, Inc, 44 Montgomery Street, Suite 250, San Francisco, CA 94104, fax: 415-646-0010, mhincks@chemconnect.com
We will present actual stories of chemical buyers and sellers who have realized the benefits of e-commerce to help them increase sales and cut costs. With the B2B e-commerce industry expected to explode, we will also provide insight into the impact of that explosion to the chemical industry and how companies can embrace the new technology to improve their business. Quoting Bloomberg News, "Major chemical companies...are moving aggressively into e-commerce, driven by the promise of sales opportunities, cost savings, and the spoils that are expected to go to the trailblazers of the new market."


Panel Discussion




Electronic Notebooks and Related Systems - State of the Technology 1
R. Lysakowski, Organizer, Presiding
8:20   Introductory Remarks
8:30 45 Catalyzing R&D automation market and product innovation
Rich Lysakowski, Jr, Executive Director, Collaborative Electronic Notebook Systems Association, 800 West Cummings Park, Suite 5400, Woburn, MA 01801, fax: 781-935-3113, rich@censa.org
Industrial R&D automation markets are small by comparison with general consumer and manufacturing markets. The R&D automation market is not really one market, but rather a large collection of small niche markets, each under $USD 1 billion in size. Since computers were introduced into R&D, these market size problems have led to slow innovation, slow standardization, and many other market inefficiencies. This is ironic given that R&D automation tools help drive innovation in the R&D process itself. A better and faster market innovation process has been needed for many years to remove the barriers and inefficiencies in the R&D automation tools market.

CENSA is a non-traditional trade association that leads multi-industry collaborations for creating advanced automation systems for R&D in leading science-oriented companies. CENSA is focused at the meta-level of market development, working to improve market innovation processes themselves. CENSA hosts exciting market development programs that are accelerating global industry's progress into 21st Century automation systems. Our current programs focus on project data handling, collaborative computing, electronic recordkeeping, and knowledge management for R&D.

CENSA End User Members, Supplier Members, and staff are working in coordinated market development programs to achieve a common vision in phases. The first CENSA-specified products have been available since March 1999; more waves of products will be delivered in 2000. Applications of these products include product research, development, manufacturing, and testing in high-throughput screening, analytical chemistry, biotechnology, healthcare, environmental and related areas.

This presentation will cover CENSA's mission, programs, strategies, and processes. Recent progress from CENSA's research, product development, and market development programs will also be presented. We will also address how the international community and regulatory agencies and patent offices are participating in and benefiting from CENSA's work.

9:00 46 Industrial-strength electronic recordkeeping and electronic notebook systems for research and development
Rich Lysakowski, Executive Director, Collaborative Electronic Notebook Systems Association, 800 West Cummings Park, Suite 5400, Woburn, MA 01801, fax: 781-935-3113, rich@censa.org
When designed and applied well, Electronic Recordkeeping Systems (ERS) and Collaborative Electronic Notebook Systems (CENS) greatly enhance global productivity and collaboration on complex R&D projects, while being reliable and trustworthy for recordkeeping from legal and regulatory points of view. When designed or applied poorly, they put intellectual property, patents, and regulatory compliance -- for companies with even the best intentions -- at serious risk, leading to bad legal exposure, very expensive litigation, damaging settlements, and lost opportunities.

Developing electronic laboratory notebook systems requires a detailed and accurate understanding of all legal, regulatory, scientific, technology, and business requirements. However, like no other lab automation systems, Collaborative Electronic Notebook Systems will have a huge impact on the practice of scientific R&D and engineering design. A unique collaboration of many Fortune 100 companies and other organizations, called the Collaborative Electronic Notebook Systems Association (CENSA) is leading the paradigm shift from paper-based recordkeeping to fully electronic recordkeeping systems. CENSA End User Members and staff are defining and prioritizing products requirements for industry at large. Then they are selecting and working with the best-qualified vendors worldwide to build products and do extensive product validation and testing. This large association of global companies and other organizations is addressing the legal and regulatory acceptance questions in partnership with government agencies worldwide.

This presentation will provide: 1) an overview of ERS and CEN systems; 2) the legal, regulatory, technical, and business imperatives that must be met to implement successful systems; 3) an overview of pilot projects in member companies to implement ERS, CENS, and related R&D automation systems.

9:30 47 e-Records, e-paper, e-books, and e-notebooks: Capturing, preserving, and accessing records for long periods of time
Marion Melani, eBooks Program Manager, Adobe Systems, 321 Park Avenue, E14, San Jose, CA 95110, fax: 408-537-4005,mailto:%20mwm@adobe.com
Electronic Books (called "eBooks") are specialized types of records that use computer technologies for the authoring, publishing, distribution, and commercial protection of rights to intellectual property. eBooks include common viewing software available on many types of reading devices, from standard PC-based to hand-held commputers. Scientists, engineers, and others who author notebooks do not think of themselves as "publishing." However, in the electronic age publishing is a natural conclusion of authoring.

Adobe has created software that allows anyone to create high-quality publications from any desktop applications, and commit them as permanent electronic records. Specialized security components (called "plug-ins") for digital signatures, time-stamping, notary services, and encryption achieve security and privacy if necessary. These permanent records can then be stored and accessed in any document management or recordkeeping system for as long as needed by an organization. ePaper, eBooks, eRecords and eNotebooks are a natural progression of technologies.

This presentation will discuss recent research and developments in ePaper and eRecords that will significantly impact the practice of science and engineering.

10:00 Intermission
10:15 48 Integrating electronic notebook systems with traditional databases
Sheila Ash, Business Development, Oxford Molecular, Medawar Centre, Oxford Science Park, Oxford, OX4 4GA, United Kingdom, fax: +44 1865 784600, sash@oxmol.co.uk, Simon Coles, NIP, Manhattan House, 140 High Street, Crowthorne, Berkshire, RS45 7AY, and John Woods, Cheminformatics, Oxford Molecular, United Kingdom.
The generation of knowledge is the primary function of every research organization. However, without appropriate mechanisms to capture and disseminate this knowledge, no commercial advantage can be taken. It was the recognition of the need to capture data about samples and reactions that brought about the success of chemical database systems. Whilst these traditional databases are well suited to storing structured information, such as assay data, we shall explain why such databases are not ideal for capturing ideas, workflows and rationales. Historically, all such contextual knowledge has been captured in the lab notebook. We shall examine the enduring popularity of the paper notebook and look at the lessons that must be taken forward into an electronic system. Finally, we shall discuss how our chemical information system has been integrated with an electronic notebook to create an electronic, chemically intelligent, lab notebook.
10:45 49 Review of the PatentPad® recording system as bridge to the electronic notebook
Tom Ryan, SCRIP-SAFE® Security Products Inc, 11319 Grooms Road, Cincinnati, OH 45242-1405, fax: (513) 489-5603
The PatentPad® recording system is a paper-based alternative to the bound research notebook. The system provides the opportunity to make better use of existing and future information management systems without compromising intellectual property issues. Described as the Bridge to the Electronic Notebook, the PatentPad® recording system has now been in full application since it was introduced to Bristol Myers Squibb in the Fall of 1997. After a brief explanation of the tool, the PatentPad® will be reviewed with respect to four research companies, Bristol-Myers Squibb Pharmaceutical Research Institute, Procter and Gamble, Purdue Pharma, L.P., and Affymax Research Institute. These four companies differ in size and structure, and each is in a different stage of maturity with respect to the tool. The review will provide insight into the value of the program as an intermediate step toward the fully electronic solution.
11:15 50 Internet authoring: The key to next generation e-notebooks
Rick Akie, Vice President, IceBreaker Communications, 1320 Centre Street, 4th Floor, Newton Center, MA 02459-2400, fax: 617.928.2982, rakie@icebreaker.com
While the Internet has provided access to vast information resources and facilitated electronic commerce, the next generation of the Internet will provide teams of people and individuals the opportunity for real-time collaboration and authoring. Collaborative authoring capabilities represent the key to next generation e-Notebooks.

This new generation of technology will allow individuals within a corporation to gather data from across an extended enterprise, and allow multi-user collaboration in a real time, web-enabled environment. In this way, organizations will transform personal data into corporate knowledge and accelerate competitive advantage.

This vision requires that e-Notebooks possess the ability to scale from personal use to full sharing and collaboration over the Internet. A well-designed e-Notebook authoring capability is the natural next step to leverage the Internet and realize the power of an e-Notebook solution.

This paper will review the importance of Internet authoring and provide the necessary e-Notebook functional and technical requirements needed to achieve break-through results.

11:45 51 Electronic recordkeeping systems: Simplified solutions over complex infrastructures
Brian MacLeod, Vice President of Product Development, PSSoftware Solutions Ltd, 1686 Woodward Drive, Ottawa, ON K2C 3R8, fax: 613-226-5447, macleodb@pssoft.com
There is a growing trend in today's market place for enhanced Web deployment of all electronic and e-mail records. Corporate end users today are asking for a fully integrated Web-based enterprise records management solution that delivers a complete range of business critical records management functionality. Companies today are generating and managing more and more electronic records, but when do these electronic documents/e-mails become records? And should corporate end users be involved in the profiling of e-mail? As e-mail begins to replace the internal office memo, it is becoming increasingly important to set standards. Moving into the 21st century, electronic records management will become a part of every business function.




Electronic Notebooks and Related Systems - State of the Technology 2
R. Lysakowski, Organizer, Presiding
1:15   Introductory Remarks
1:30 52 Tool to collect scientific data using a corporate intranet
Mark Woodford, Laboratory Automation Group, Monsanto Life Sciences, 4901 Searle Parkway, Skokie, IL 60076, mark.m.woodford@monsanto.com
There are many thousands of scientific devices at Monsanto - at several research sites- in half a dozen states. The talk will relate the experience of the Monsanto Laboratory Automation Group in using the intranet to collect diverse data streams from 45 geographically disparate scientific devices at research sites in Missouri, California, and Illinois using a Monsanto-designed data acquisition server. This key technology was written by Peter Timov of the Monsanto Laboratory Automation Group.

Based on our experience, this talk will relate the potential benefits of being able to touch large numbers of diverse scientific and medical instrument communication streams from a central source and will describe the technical, organizational, and financial barriers surrounding the ability to achieve these connections.

The potential benefits of this technology toward the creation of common scientific device control functions will be presented. The talk will also describe industry association, standard group, and commercial initiatives for collecting scientific data streams. Much of what is presented will be technical in nature - describing and differentiating networking protocols, OSI model layers, and how they relate to communication between scientific devices.

2:00 53 Web-based electronic laboratory notebook developed by Battelle Pacific Northwest Laboratories.
Elena Mendoza, Environmental Molecular Sciences Laboratory, Battelle Pacific Northwest Laboratories, Battelle Blvd. MS K1-87, Richland, WA 99352, elena.mendoza@pnl.gov and James D. Myers, Collaboratory Group Leader, Battelle / Pacific Northwest National Laboratory, Battelle Blvd. MS K1-87, Richland, WA 99352.
As part of Pacific Northwest National Laboratory's Environmental and Molecular Sciences Collaboratory project and the Department of Energy's DOE2000 Electronic Notebook project, we have developed a WWW-based Electronic Laboratory Notebook (ELN). The ELN is a multi-media, sharable, and interactive version of the traditional paper laboratory notebook. It allows user to share data, sketches, molecular structures, equations, etc. and can easily be extended to support additional types of entries. The ELN also supports encryption and digital signatures, searching, automated entry from instruments, and other advanced capabilities. The Electronic Notebook is currently used by a variety of researchers and educators around the country, and is a standard service offered to users of the Environmental Molecular Sciences Laboratory (EMSL). In education it allows students to share their data and experiments and get feedback from their instructors and fellow students. Also, interns and student researchers are able to connect with their research sponsors in industry in separate locations. In this talk we will discuss the features of the Electronic Laboratory Notebook and present details of how several groups have used it to support their research and education projects.
2:30 54 Case studies: The implementation of electronic laboratory notebooks for synthetic chemistry
Keith T. Taylor, Product Marketing, MDL Information Systems Inc, 14 Walsh Drive, Parsippany, NJ 07054, fax: 973-299-3995, keitht@mdli.com and Philip McHale, Product Marketing, MDL, 14600 Catalina Street, San Leandro, CA 94577.
Before an electronic notebook system is accepted, it is essential that it does not compromise the company’s intellectual property and chemists must be willing to use it. The technologies required to ensure the protection of intellectual property have now matured. During the past five years MDL has developed a number of electronic laboratory notebooks for use by synthetic organic chemists. Early approaches used a forms based user interface, which many users find inflexible. Today we are able to implement a more familiar and flexible document based user interface. The evolution of the electronic laboratory notebook will be illustrated with examples from actual systems. Current systems capture most of the information types that are required but with few exceptions the data is static. The electronic notebook will not reach its full potential until users can interact with captured data. Developments that facilitate the fully interactive electronic laboratory notebook will be discussed.
3:00 55 ProGen: An approach for the collection, annotation, and structuring of experimental information
Mike Mochan, Technical Marketing, Medical Sciences Associates, 6565 Penn Avenue, Pittsburgh, PA
Management Science Associates, Inc., (MSA) has built a system for the capture and annotation of experimental data as it is generated throughout the design, implementation and analysis of a study or procedure. This system (ProGen) was designed using flow cytometry as a model, however we believe our general approach is adaptable to other laboratory processes and applications. Key to this concept is the reusability of eight core modules of the program. These modules address the following areas:

1. Organizational, Study, and Experimental Configuration. 2. Experiment Definition, Procedures and Instructions. 3. Experiment Setup and Execution Control. 4. Laboratory Device and Data Acquisition Setup. 5. Chemical Formulae Definition and Control (Reagent Cocktail). 6. Laboratory Material Preparation (Staining). 7. Sample, Specimen, and Subject Management. 8. Laboratory Inventory Control.

ProGen provides a paperless system that fulfills the role of an ELN by capturing and annotating the information surrounding the experiment's design and execution. The annotated data provides a permanent means of documenting the context in which an experiment is conducted. By storing this data in an electronic format, it can easily be recalled for subsequent analysis or can be used to repeat or verify an experiment.

ProGen is based on a scalable architecture and is written in JAVA to run on a wide variety of computer hardware platforms. It uses a relational database to store and structure information regarding studies, experiments, protocols, stains, procedures, subjects and specimens. ProGen is also designed to support day-to-day laboratory operational and management functions (e.g. reagent inventory tracking, work-list generation and staffing workload). ProGen utilizes the extensible markup language (XML) to communicate methodology and reagent information to machine devices. Annotation data and raw output data are then combined with analytical information using XML notation to form a file set that is suitable for permanent archive and retrieval. The XML file contains keywords that facilitate loading into a lightweight directory access protocol (LDAP) database for publishing flow cytometry data on the World Wide Web.

3:30 Intermission
3:45 56 Documentum and electronic recordkeeping
Bob Little, iResearch Application Suite Product Management, Documentum, Inc, 6801 Koll Center Parkway, Pleasanton, CA 94566, fax: (925) 600-6850,mailto:robert.little@documentum.com
The US Food and Drug Administration, through the Code of Federal Regulations, Title 21, Food and Drugs, part 11, "Electronic Records; Electronic Signatures" (21CFR11), has issued requirements for electronic signatures and record management. These regulations are having a profound impact on the manner in which organizations are handling electronic records. In addition, these requirements are influencing other government agencies as they address similar issues.

To enable conformity with these requirements, Documentum has developed software that allows organizations to electronically sign and securely audit documents within a content management infrastructure. This session will review the regulatory landscape and will discuss how Documentum can be utilized to achieve compliance. Topics to be discussed include 21CFR11 requirements, user authentication, electronic signing of documents, audit trails, record management and administration, integration with applications, and implementation issues.

4:15 57 Panel Discussion. Implementation successes and issues
Rich Lysakowski, Executive Director, Collaborative Electronic Notebook Systems Association, 800 West Cummings Park, Suite 5400, Woburn, MA 01801, fax: 781-935-3113, rich@censa.org
The Symposium on Electronic Notebooks and Related Automation Systems will conclude with a moderated roundtable discussion with all the day's speakers present to participate, answer questions from the audience, and prognosticate on automation developments in the near future.




Modeling and Informatics for Nonexperts
O. Guner, Organizer, Presiding
8:25   Introductory Remarks
8:30 58 Modeling for experimental chemists: Design principles/strong>
George D. Purvis III, Oxford Molecular, 14940 NW, Greenbrier Pkwy, OR 97007, fax: 503-533-5099, gpurvis@oxmol.com
CAChe implements novel human interface design principles to achieve ease of use and high productivity from non-experts. Principles for human interface design are explained and examples displayed covering ease of use, econoomical use of time through automotation, and information tracking. Finally, the advantages of software tailored to a specific application vs general purpose software are discussed with CAChe for Medicinal Chemists used to illustrate key points.
9:00 59 Keeping it simple: Designing intranet-based chemical information systems for nonexpert users
Glenn Atter, David Saul, and John Woods. Cheminformatics, Oxford Molecular, Medawar Centre, Oxford Science Park, Oxford, OX4 4GA, United Kingdom, gatter@oxmol.co.uk
Chemical information systems are difficult for nonexpert users. Because of this, data can gather dust in "write-only" corporate databases rather than help scientists to increase their research productivity. The various barriers that prevent nonexperts from making productive use of chemical information systems will be described. Overcoming these barriers is a key issue for software developers and IT professionals. An approach that uses the corporate Intranet to deploy chemical information applications specifically for nonexpert users will be discussed. Examples relating to the handling of chemical and biological data in pharmaceutical discovery will be used to show how applications can be designed to break down barriers to access and usability, and to illustrate how this approach enables the nonexpert user to make productive use of chemical information.

Presentation (pdf)

9:30 60

Informatics at arena pharmaceuticals: An integrated web-based system
Gareth Jones, Arena Pharmaceuticals, Inc, 6166 Nancy Ridge Drive, San Diego, CA 92121, fax: 858-453-7210, gjones@arenapharm.com
The development of platform-independent web-based computing allows ordinary users unprecedented access to corporate information. At Arena Pharmaceuticals we have developed a web-based informatics system that allows all employees access to chemical, screening, genomic and gene-expression data. This system was designed specifically to allow users with little or no computing experience the ability to browse, analyze, update and edit chemical and biological data. This results in real-time distribution of experimental data and allows on the fly analysis and search of information. Additionally, communication between various groups (e.g. Chemists and Biologists) working on the same project has been greatly facilitated. The data system is based on Oracle, Unity and Blast databases. Data is published and updated via a webserver with CGI programs and client Java software. Using a secure socket layer (SSL), the database has been made available to corporate partners. The use of authentication protocols and Oracle roles allows collaborators instant access to relevant data in a secure environment. The availability of timely information is invaluable, especially when a collaboration spans many locations and timezones.

10:00 61 Integration issues in pharmaceutical informatics: The interface between informatics, chemistry, and molecular modeling for experimental scientists
H.J.R. Weintraub, Genentech, Inc, 1 DNA Way, MS 56A, South San Francisco, CA 94080, fax: (650) 225-1356, hjrw@gene.com
With the maturing of the technologies associated with combinatorial chemistry and parallel synthesis, high throughput screening (HTS), and rapid analytical profiling of chemical libraries there is a pressing requirement to develop integrated approaches to pharmaceutical informatics. This approach is necessitated by the business need to efficiently analyze all available data as rapidly as possible to facilitate decision making, and to provide required information for regulatory agency filings. This presentation will focus on an integrated approach to information management. The work flow to be discussed ranges from chemical library design through diversity estimation, plate layout consideration, association of analytical and HTS data, and total integration of molecular modeling analysis. The development of a model informatics system using off-the-shelf and custom in-house-developed software components is described.
10:30 62 Simulation of infrared spectra on the web
Johann Gasteiger, Markus Hemmer, and Paul Selzer. Computer-Chemie-Centrum, University of Erlangen-Nurnberg, Nagelsbachstrasse 25, Erlangen, D-91052, Germany, fax: +49-9131-85 26566, Gasteiger@ccc.chemie.uni-erlangen.de
A novel coding of the 3D structure of molecules combined with the powerful modeling technique of a counterpropagation (CPG) neural network allows the simulation of high quality infrared spectra. This approach is now offered through the TeleSpec project on the internet. The user enters the 2D structure of a molecule and obtains an infrared spectrum proposed for this compound. The mathematical background and the various steps in this method - generation of a 3D structure, encoding of this structure, training of a CPG network, prediction of an infrared spectrum by this neural network - are explained and visualized on the WWW. The approach can also be turned around and allows the prediction of a 3D structure of a molecule directly from its infrared spectrum. A collaborative effort has been established to build a database of infrared spectra for the academic community.
11:00 63 Did I make what I thought I made? IR and NMR simulation in compound verification
Gregory M. Banik, Sadtler Division, Bio-Rad Laboratories, 3316 Spring Garden Street, Philadelphia, PA 19104, fax: 215-662-0585, gregory_banik@bio-rad.com
Analytical techniques such as IR and NMR spectroscopy are standard tools in verifying the identity of novel compounds prepared by synthetic chemists. The use of software simulation tools to simulate IR and NMR spectra and assist in the verification process will be described.
11:30 64 Novel methods for visual interpretation of biological screening data
Glenn J. Myatt, Kevin P. Cross, Paul Blower, Wayne P. Johnson, and Gulsevin Roberts. Research and Development, Columbus Molecular Software, Inc, Business Technology Center, 1275 Kinnear Rd, Columbus, OH 43212, fax: 614-675-3732, gmyatt@columbus-molecular.com
Modern approaches to drug discovery have dramatically increased the speed and quantity of compounds created and tested for potential potency. The task of organizing, and interpreting this information is a major bottleneck in discovering new drugs. We will discuss how LeadScopeTM uses novel, interactive visualizations for browsing and interpreting chemical and biological screening data to assist pharmaceutical scientists in finding promising drug candidates. The software organizes the compounds using structural features and statistics familiar to medicinal chemists. Different graphical methods visually summarize the data; highlighting common structural classes and custom features that are statistically correlated with biological activity and property data.




General Papers
A. Berks, Presiding
1:00 65 Chemical information systems for mixtures, formulations, and other composite substances
David Benham, Dale Gray, and John Woods. Cheminformatics, Oxford Molecular, 11350 McCormick Road, Executive Plaza III, Suite 1100, Hunt Valley, MD 21030, dbenham@oxmol.com
Substances such as detergents, natural products, poorly-characterized chemical mixtures and other mixed materials present a more complex challenge for chemical information systems than pure, well-characterized chemical substances. The next-generation chemical information systems of the new century will be able to handle such materials and thus become applicable to new scientific areas. As well as the ability to specify and store a wide range of mixtures, a chemical information system must also allow a user to search for mixed substances in intelligent ways - for example, finding all formulations that contain a mass fraction of a given material that is within a specified range. The challenges involved in creating such a system will be described along with possible solutions.
1:30 66 MolMall (www.molmall.org) nonprofit project for chemical exchange
Luc Patiny, Institute of Organic Chemistry, University Lausanne, ICO-BCH, 1015 Lausanne, Lausanne, 1015, Switzerland, fax: +41 21 692 39 55, Luc.Patiny@ico.unil.ch and Shu-Kun Lin, MDPI Center, Molecular Diversity Preservation International, Sanegergasse 25, Basel, CH-4054, Switzerland.
The nonprofit international organization MDPI launched the first fully web-based, state-of-the-art virtual market for chemical exchange with substructure searchable, integrated database for chemical reagents from chemical companies (e.g., Aldrich) and rare chemical samples from individual chemists, at http//www.molmall.org server. The software from ChemExper has been already successfully used for publishing and substructure searching MolBank section of the online journal Molecules at the http://www.molbank.org server. Many chemistry journals (e.g., Acta Chemica Scandinavica) started to encourage authors to deposit samples at MDPI in Switzerland at their instruction for authors. Authors can submit structures of samples directly on the internet. Chemist can search for reagents and rare samples on the constantly updated database with one query of substructure or structure, or contributors (suppliers), IUPAC name, etc. or their combinations. The result displays as table on the internet. Progresses and some planned improvements will be discussed.
2:00 67 Systematic naming technology
Carlton G. Squires, Advanced Chemistry Development, 133 Richmond St., W, Suite 605, Toronto, ON M5H 2L3 Canada, fax: 416-368-5596, carlton@acdlabs.com
Several strategies have been developed in recent years to automate the process of systematic chemical name generation. Though a few techniques have shown success in certain areas of chemistry, in many cases, prior attempts have led to an overall level of accuracy that has been less than stellar. Through algorithms developed and refined over the past few years, a procedure for systematic name generation with a focus on accuracy will be demonstrated. These algorithms have been successfully applied to generation of systematic names according to IUPAC and CAS Index recommendations for almost any organic structure, and selected classes of biochemical, organometallic and inorganic structures. Furthermore, the growth of electronic media and searching makes recent advances in accurate name-to-structure capability extremely timely. These name-to-structure capabilities will also be demonstrated. Whether implemented together or separately, these algorithms provide a method for unambiguous generation and communication of systematic chemical names and their structures.
2:30 68 Introduction to the chemical registry system at the Environmental Protection Agency
Tommie G. Curtis and Lois E. Fritts. Science Applications International Corporation, 6565 Arlington Boulevard, Suite 100, Falls Church, VA 22042-3000, curtist@sdc-moses.com
The Environmental Protection Agency (EPA) has developed a Chemical Registry System (CRS) that is intended to be the central source of chemical identification information for the Agency. The system is available to EPA staff and the public on the Agency web site (http://www.epa.gov/crs/). In the CRS, a substance can be queried by Chemical Abstract Service Registry Number (CASRN), systematic name, synonym, or regulation. Search results include detailed chemical identification information, the regulations that list a chemical, chemicals in a specific regulation, and the Agency office responsible for a regulation or information system that includes the substance. The system also provides downloadable domain value lists usable in the development of new or reengineered systems.
3:00 69 Reinventing chemical information at the Environmental Protection Agency
Lois E. Fritts and Tommie G. Curtis. Science Applications International Corporation, 6565 Arlington Boulevard, Suite 100, Falls Church, VA 22042-3000, frittsl@sdc-moses.com
EPA has historically responded to Congress by creating independent offices to enforce legislation. Each office devised a schema for identifying chemicals: Water created parameter codes; Air, pollution codes; Solid Waste, hazardous waste codes. No process existed for aggregating data about a chemical across media. In response to programs such as Reinventing Environmental Information, the Agency established the Environmental Data Registry (EDR) as a data management tool. Data standards are recorded in the EDR and used to develop new or reengineered systems. The Chemical Data Standards Working Group, composed of chemists representing Agency programs, has worked with the Office of Information Resources Management (OIRM) and state partners to establish a chemical data standard. The key data elements, valid domains of chemical identifiers, and business practices of the chemical data standard are documented in the EDR.