Track 5 explores partnerships and collaborations to further drug discovery, open source chemistry, recent advances in cheminformatics, modeling for safety, using biological and chemical information to guide hit-to-lead phase and lead optimization, and repurposing drugs by applying 21st century tools to find new targets.

TUESDAY, APRIL 20

2:00 - 6:00 pm Main Conference Registration

4:00 Event Chairperson’s Opening Remarks

Cindy Crowninshield, Conference Director, Cambridge Healthtech Institute

Presented by
4:05  Keynote Introduction: Ronald Ranauro, Chief Executive Officer, GenomeQuest, Inc. 

4:15 PLENARY KEYNOTE: Drug Discovery Opportunities and Challenges -- VC, Biotech and Pharma Perspectives Christoph Westphal, M.D., Ph.D., CEO, Sitris Pharmaceuticals; Senior Vice President, Center of Excellence for External Drug Discovery, GlaxoSmithKline

 

5:00 - 7:00 pm Welcome Reception in the Exhibit Hall

***Drop off your business card at the CHI Sales Booth for a chance to win an Apple® - iPod nano®!  2 Winners will be announced at 6:45pm in the Exhibit Hall

 

Wednesday, April 21

7:30 am Registration and Morning Coffee

Sponsored by
 8:15 Event Chairperson’s Opening Remarks

Phillips Kuhl, Co-founder and President, Cambridge Healthtech Institute

Keynote Introduction: Jamie Wyatt, Vice President and General Manager Health and Life Sciences, Netezza

8:20 PLENARY KEYNOTE: Impact of HIT Stimulus on Novel Sources of Data for Research John Halamka, M.D., M.S., CIO, Harvard Medical School           9:00 Keynote Presentation & 2010 Benjamin Franklin Award  Alex Bateman, Ph.D., Senior Investigator, Pfam Database Project, Wellcome Trust Sanger Institute

 

Sponsored by
 9:30 Coffee Break, Poster and Exhibit Viewing

 ***Drop off your business card at the CHI Sales Booth for a chance to win 1 of 2 Prizes! (Nintendo® Wii™ System or Apple® - iPod touch®)

 

Partnerships And Collaborations:
A New Paradigm In Drug Discovery

10:50 Track Chairperson’s Remarks

John Overington, Ph.D., Team Leader, Computational Chemical Biology, EMBL-EBI

11:00 Partnerships and Collaborations: A New Paradigm in Drug Discovery

Barbara Mittleman, M.D., Director, Program on Public-Private Partnerships, Office of Science Policy, NIH

Dr. Mittleman will discuss how the NIH views and implements partnerships with entities in industry, in the non-profit sector, and with other government agencies to accomplish the NIH’s mission to improve the public health through biomedical research. Drug development is an important component of accomplishing our mission. Partnerships focused on developing and qualifying biomarkers, in developing high throughput screening methods, in identifying novel disease targets and development schemata, etc. will be discussed in the setting of both pre-competitive and competitive approaches.

11:30 Precompetitive Collaborative Opportunities in Cheminformatics

Ramesh Durvasula, Ph.D., Director, Chemistry Informatics, Research Informatics & Automation, Bristol-Myers Squibb

In early 2008, work began in earnest on the creation of a precompetitive collaboration legal construct, the Pistoia Alliance. Initially targeted toward the identification of cross-pharma opportunities in chemistry informatics (common processes, data models, etc.), the focus rapidly expanded to include all components of the research workflow as the group expanded. Currently, there are active work streams spanning chemistry, biology, and knowledge management. This talk will review past, present, and emerging pre-competitive cheminformatics opportunities within and outside the Pistoia Alliance construct.

12:00 pm Scientific Informatics Outsourcing – New Trend in the Life Sciences Industry

Richard Golob, CEO, GGA Software Services LLC

The historical paradigm in the life sciences industry has been to use outsourcing companies primarily for non-scientific informatics activities. However, with the advent of outsourcing companies that have strong scientific and mathematical expertise, this paradigm is shifting towards scientific informatics outsourcing, from software engineering, algorithm development, and knowledge management to testing, support and maintenance, and scientific BPO.  Through scientific informatics outsourcing, life science companies have an opportunity to extend their workbench in a cost-effective way and to complement and supplement their internal teams.

12:30 Luncheon Presentation (Sponsorship Opportunity Available) or Lunch on Your Own

 

Open Access Chemistry And Biology

1:40 Chairperson’s Remarks

Antony Williams, Ph.D., FRSC,VP, Strategic Development, ChemSpider, Royal Society of Chemistry

1:45 Linking Biology and Chemistry Spaces - Prospects for Improved Drug Discovery and Development

John Overington, Ph.D., Team Leader, Computational Chemical Biology, EMBL-EBI

Large-scale databases of small molecule and target interactions can now be analyzed to uncover fundamental patterns and constraints in drug-like and druggable target space. These rules can be used to prioritize likely successful targets, to identify compounds for screening studies, and also provide the basis for data integration tools for translational drug discovery.

2:15 Chemical Information Mining Today & Tomorrow

Debra Banville, Ph.D., Information Analyst, Discovery Information, AstraZeneca Pharmaceuticals

Chemical information mining is rapidly developing in the public sector as a way to find and share key information about chemical entities within a broad community. This talk will focus on the latest developments and project into the future the potential paradigm shift we could expect. This presentation will give the audience an update on what’s currently available and will offer insight into how these developments will change the way we search and interact with scientific literature in general.

2:45 Machine Learning Approaches for Molecular Modeling

Ramgopal Mettu, Assistant Professor, Electrical and Computer Engineering, UMass Amherst

Often in molecular modeling we wish to study a particular target under a number of different simulation conditions (e.g. testing a database of ligands, or probing for binding sites). In recent years, mining and inference techniques from the machine learning community have become widely used for molecular modeling. In this talk, I will give a overview of the technical ideas behind these methods, and discuss their application to ligand-binding prediction, computational mutagenesis, and protein design.

3:15 Refreshment Break, Poster and Exhibit Viewing

***Drop off your business card at the CHI Sales Booth for a chance to win 1 of 2 Prizes! (Nintendo® Wii™ System or Apple® - iPod touch®)  Winners will be announced at 3:30pm in the Exhibit Hall

Recent Advances in
Cheminformatics and Screening

3:45 Knowledge-Based Compound Perturbation of Biological Networks

Meir Glick, Ph.D., Research Investigator II, Center for Proteomic Chemistry, Novartis

Literature-curated networks of biological interactions are traditionally used to expand our understanding of cellular and pathway biology. In this work, networks are used as the organizing principle for target and compound selection to create focused screening libraries. We demonstrate the use of network-focused libraries in assay development, tool compound selection, lead discovery, and potential chemical combinations.

4:15 Integration with PubChem Biological Activity Reporting on Small Molecules for Better Project Decision Making

Raza Shaikh, Associate Director, Informatics, Chemical Biology Platform, The Broad Institute

We will present how a programmatic integration with PubChem compound, assay and activity data has enabled faster and better decision making on which hits to take forward in a small molecule probe development campaign. There is a wealth of data being deposited in PubChem and manually browsing through is really time consuming and cumbersome. A unique pivot report of compounds against known assays for biological activity that is generated in real-time on your hits allows this process to be efficient and effective.

4:45 Bioassay Ontology and Software Tools to Integrate and Analyze Diverse Data Sets

Stephan Schurer, Ph.D., Assistant Professor, Pharmacology and Center for Computational Sciences, University of Miami

Increasingly large and diverse data sets are being generated by publically funded screening centers using various high- and low-throughput screening technologies. Much of this data is accessible, for example via PubChem, the largest public repository of small molecule screening results, currently covering over 1,500 assays for 370,000 compounds. The number of assays can be expected to grow more than 10 fold during the next five years. The utility of this invaluable resource is currently limited, because the knowledge contained in complex and diverse bioassay data sets is not  formalized and therefore cannot be accessed for comprehensive computational  analysis or integration with other data sources. This proposal is to attack  this limitation. For the past ten years ontologies have been developed by  biologists to facilitate the analysis and discussion of the massive amounts  of information emerging from the various genome projects. An ontology is a  controlled vocabulary representation of the objects and concepts and their  properties and relationships. The purpose is to model and share  domain-specific knowledge so that software agents can automatically extract  and associate information. The aim of this proposal is to develop a bioassay  ontology, software tools, and to demonstrate their utility. The bioassay  ontology will coherently describe diverse biological assays (such as those  in PubChem) with a focus on complex cell-based assays and in particular  high-content screening data. Software support and development includes  modules to build ontology terms and to curate data sets, tools to map the  ontology onto screening experiments and other ontologies, and tools to  standardize, reformat, and aggregate data sets in the context of the  ontology. We will demonstrate the utility of our approach by creating a  PubChemderived database and making it available to the community via a  search interface. The ontology and software tools will facilitate the  analysis of bioassay screening data in various contexts, for example  signaling or metabolic pathways and indirectly human disease. The tools will  enable one to extract data sets for modeling specific interactions between  perturbing agents and biological targets (or pathways), or to model assay  technology-dependent interferences. End user software needs to provide ease  of use for biologists and chemical biologists to utilize the ontology in the  context of their own and external data sets. It will be modular and open  source. We will develop various collaborations to disseminate the bioassay  ontology and software in the community and to facilitate their ongoing  development. 

5:15 – 6:15 2010 Best of Show Awards in the Exhibit Hall

6:15 Exhibit Hall Closes

Sponsored by
6:30 – 10:00 2010 Best Practices Awards Reception & Dinner

 

 

 

 

 

THURSDAY, APRIL 22

8:00 am Registration and Morning Coffee

Sponsored by
8:45 Event Chairperson’s Opening Remarks

Kevin Davies, Ph.D., Editor-in-Chief, Bio-IT World

Keynote Introduction:  Eric Blatte, Vice President of Sales, Commercial & Public Sector, Imprivata

8:50 PLENARY KEYNOTE: There is No Magic, There is Only Awesome: Scientific Computing with Amazon Web Services Deepak Singh, Ph.D., Business Development Manager, Amazon Web Services   Presentation delivered via a live, interactive videoconferencing platform.     9:30 KEYNOTE PANEL The Future of Personal Genomics A special plenary panel discussion featuring: James Heywood, Co-founder and Chairman, PatientsLikeMe Dan Vorhaus, J.D., M.A., Attorney, Robinson, Bradshaw & Hinson; Editor, Genomics Law Report Dietrich Stephan, Ph.D., President & CEO, Ignite Institute Kári Stefánsson, MD, Dr Med, Executive Chairman and President of Research, deCODE genetics Kevin Davies, Ph.D., Editor-in-Chief, Bio-IT World

 

10:30 Coffee Break, Poster Competition, Vendor Theater Presentations and Exhibit Viewing

***Drop off your business card at the CHI Sales Booth for a chance to win 1 of 2 Prizes! (Nintendo® Wii™ System or Apple® - iPod touch®)

Structure-Informed Modeling

10:55 Track Chairperson’s Remarks

Tudor I. Oprea, Ph.D., Professor & Chief, Biochemistry & Molecular Biology Biocomputing, Health Sciences Center, University of New Mexico

11:00 Improving Proteochemometrics Modeling for GPCRs and HIV Reverse Transcriptase Using Protein Fingerprints

Andreas Bender, Ph.D., Assistant Professor for Cheminformatics and Pharmaceutical IT, Medicinal Chemistry Division, Leiden/Amsterdam Center for Drug Research

Proteochemometrics Modeling is a recently introduced valuable technique in computer-aided drug design that uses both ligand and target features to predict ligand activity also against novel receptor subtypes or protein mutants. In this work we describe how rationally useful protein/amino acid descriptors can be chosen, and we present prospective validations on selecting active ligands of G-Protein Coupled Receptors and HIV Reverse Transcriptase. Given that proteochemometrics models more fully exploit information given in ligand-target bioactivity matrices than conventional QSAR models, this leads to improved predictive models as we will show here.

11:30 PRDB: A Protein Relational Database and Protein Family Knowledge Bases to Facilitate Structure-Based Design Analyses

Dominick Mobilio, Ph.D., Director, Cheminformatics, Pfizer, Inc.

The Protein Data Bank is the most comprehensive source of experimental structures. However, it can be difficult to locate relevant structures with its search interface, particularly when searching for protein–ligand complexes containing specific intermolecular interactions. We describe three new databases, Protein Relational Database (PRDB), Kinase Knowledge Base and Matrix Metalloproteinase Knowledge Base, containing protein structures from the PDB. In PRDB, atom-atom distances between protein and ligand are pre-calculated allowing for retrieval based on these data in less than one second.

12:00 pm  Predicting Multiple Ligand Binding Modes Using Self-Consistent
Pharmacophore Hypotheses

Ryan Lilien, Ph.D., Assistant Professor, Department of Computer Science and Banting & Best Department of Medical Research, Faculty of Medicine, University of Toronto

We present a step toward improving protein-ligand binding mode prediction for a set of ligands known to interact with a common protein. There is thus an important distinction between this work and traditional virtual screening algorithms. Whereas traditional approaches attempt to identify binding ligands from a large database of available compounds, our approach aims to more accurately predict the binding mode for a set of ligands which are already known to bind the target protein. The approach is based on the hypothesis that each active site contains a set of interaction points which binding ligands tend to exploit. In a more traditional context, these interaction points make up a pharmacophoric map. Our algorithm first performs traditional protein-ligand docking for each known binder. The ranked lists of candidate binding modes are then evaluated to identify a set of poses maximally self-consistent with respect to a pharmacophoric map generated from the same poses. We have extensively demonstrated the application of the algorithm to four protein systems (thrombin, cyclin-dependent kinase 2, dihydrofolate reductase, and HIV-1 protease) and attained predictions with an average RMSD < 2.5 A for all tested systems. This represents a typical improvement of 0.5-1.0 A (up to 25%) RMSD over the naive virtual docking predictions. Our algorithm is independent of the docking method and may significantly improve binding mode prediction of virtual docking experiments.

12:30 Luncheon in the Exhibit Hall

***Drop off your business card at the CHI Sales Booth for a chance to win 1 of 2 Prizes! (Nintendo® Wii™ System or Apple® - iPod touch®)  Winners will be announced at 1:45pm in the Exhibit Hall

2:00 Exhibit Hall Closes

 

Applying 21st Century Tools To
Find New Targets For Old Compounds

1:55 Track Chairperson’s Remarks

Andreas Bender, Ph.D., Assistant Professor for Cheminformatics and Pharmaceutical IT, Medicinal Chemistry Division, Leiden/Amsterdam Center for Drug Research

2:00 Pharmacography: The Art of Mapping Drug-Target Interactions

Tudor I. Oprea, Ph.D., Professor & Chief, Biochemistry & Molecular Biology Biocomputing, Health Sciences Center, University of New Mexico

Our understanding of drug-target interactions has been limited by the amount of experimental evidence, and by our ability to attribute drug-target interactions to clinical relevance. Dubbed polypharmacology or secondary pharmacology, this emerging area challenges our notions of drugs selectivity. We will place the relationship between documented drug-target interactions and clinical relevance, within the context of several pharmacokinetic parameters such as MRTD (the maximum recommended therapeutic dose), plasma protein binding, and availability. Extensions to a CNS-centric model of ligand-protein interactions will be discussed as will some of our preliminary results and conclusions. Some of these efforts may serve as the rational basis for drug repurposing, i.e., identifying novel modes of action and therapeutic applications for already-approved drugs, and may have a profound effect on the way drug design will be conducted in the future.

2:30 Predicting Drug Off-Targets

Michael Keiser, Ph.D., Postdoctoral Scholar, Pharmaceutical Chemistry, University of California San Francisco

Whereas drugs are intended to be selective, at least some bind to several physiologic targets, explaining both side effects and efficacy. We compared ~3,700 drugs against hundreds of targets, defining each target by its ligands, using the Similarity Ensemble Approach. Chemical similarities between drugs and ligand sets predicted thousands of unanticipated associations. Thirty were tested experimentally, including GPCR off-targets of the transporter inhibitor Prozac and the enzyme inhibitor Rescriptor. Overall, 23 new drug-target associations were confirmed, five below 100 nM.

3:00 Recent Results in Network Pharmacology

Philip E. Bourne, Ph.D., Professor of Pharmacology, Associate Director Protein Data Bank, Editor-in-Chief, PLoS Computational Biology, University of California San Diego, Department of Pharmacology

We have developed a strategy for determining off-targets to a number of major pharmaceuticals. Finding off-targets presents the possibility of better understanding side effects, repositioning drugs and better defining dirty drugs. Beyond the targets we infer phenotypic effect through both static and dynamic network analysis. Recent results will be presented for a multi-target strategy in treating TB and in offering reasons for the failure of the CETP inhibitor Torcetrapib.

3:30 Case Study: Is There a Link Between Structure-Rich Information and Improving Potency and Selectivity?

Jose Duca, Senior Principal Scientist, 3D - Drug Design Department, Schering Plough

Serine and threonine kinases play an important role in signal transduction pathways. Within this kinase family, cyclin-dependent 2 kinase (CDK2) is an attractive oncology target involved in cell cycle regulation. In recent years, kinase inhibition has become a major area for therapeutic involvement. As we discuss here, these efforts have resulted in a considerable increase in the number of available high resolution structures of CDK2-inhibitor complexes. A large amount of structural-based and computational work has allowed identifying novel chemical scaffolds and structural motifs to design potent CDK2 inhibitors. Of any kinase, CDK2 has the most structures available from the Protein Data Bank (PDB), averaging 22 new structures per year since 2002. A protein-ligand interaction fingerprint analysis of the available CDK2 protein-ligand complexes indicates that structural diversity is attainable from structure-based design of CDK2 inhibitors. Since the first CDK2 structure was published on 1996, seven new chemical entities (NCE) have been advanced to clinical stages. To date, only three of these NCE have their complexes published in the PDB. This review summarizes the structurally-informed efforts in the field of CDK2 inhibitor design.

4:00 Conference Adjourns

Apple® - iPod touch®,  Nano®, Nintendo® Wii™, are  not sponsors or participants in this program.