Register by September 14th and SAVE up to $200!

Friday, October 12

8:00 am Morning Coffee Breakfast Workshop (Sponsorship Available) or Breakfast on your own

8:30 Chairperson’s Remarks
Mark Murcko, Ph.D.

8:35 The Wisdom of Crowds versus the Inevitability of Black Swans: What we Can and Cannot Learn from History
	Mark Murcko, Ph.D., Vice President and Chief Technical Officer, Vertex Pharmaceuticals Inc.
Within narrowly-defined ranges of molecular structure it is now possible to predict many physical and biological properties with reasonable accuracy. Good experimental data, appropriate parameterization, and careful statistical treatments are of course required, but with those in hand, the results these days are quite encouraging. More interesting, however, is the question of what we cannot predict -- both the "known unknowns" and the infamous "unknown unknowns". This is especially relevant in light of many recent events: the opening up of the genome, the renewed interest in natural products, and the desire to broaden our drug discovery horizons to include such hitherto "off limits" topics as protein-protein interactions. How can we build a compound profiling roadmap that will guide us into the coming decades?

9:15 Rethinking Similarity & the Chemotype Concept: Using Biological Fingerprints as a More Relevant Descriptor
	Jonathan S. Mason, Ph.D., Divisional Director, Early Lead Generation and Computational Chemistry, Lundbeck Research DK
A major challenge in the drug discovery process is the selection of leads and development candidates that are differentiated from other compounds with attrition risk managed. Results will be presented from a major study of drug, attrited and project compounds using systematic biological profiling (Cerep BioPrint?) and a large scale biological data integration project: The critical polypharmacology, both desired and undesired, of many drugs Differentiation by biological profile – a more relevant descriptor? Dangers of using chemical structural similarity or diversity as a selection approach  Structure-in vitro-in vivo & target-target inter-relationships

9:45 Black Swans and White Tablets
	Tudor I. Oprea, Ph.D., Professor & Chief, Biochemistry & Molecular Biology Biocomputing, University of New Mexico School of Medicine
The New Mexico Molecular Libraries Screening Center (screening.health.unm.edu) integrates virtual screening with high-throughput flow cytometry (HyperCyt™). Two successful developments, for the Formyl Peptide Receptor and for GPR30, both G-protein coupled receptors, will be highlighted. Chemical probe analyses in PubChem and WOMBAT, in the context of property profile optimization, will be discussed.

10:15 Poster Session, Exhibit Viewing, and Coffee Break

10:45 Prioritizing Hits from Phenotypic Assays Using Biochemical Profiling Data
	Jeffrey J. Sutherland, Ph.D., Computational Scientist, Discovery Informatics, Lilly Indianapolis
Multi-parameter phenotypic assays are useful for monitoring cell-cycle inhibition by small molecules. While a traditional chemical structure-based analysis of hits can be informative, the use of biochemical profiling data allows for a semi-mechanistic analysis of active molecules. To reduce cycle times, validated predictive models can be used to supplement experimentally-determined profiling data. Combined with the characterization of reference compounds, phenotypic and biochemical activity fingerprints provides useful means of understanding inhibition of the G2/M cell-cycle transition.

Case Study: 11:15 Chemogenomic Strategies for GPCR Hit Finding  Wolfgang Guba, Ph.D., Group Leader Cheminformatics & Molecular Modeling, F. Hoffmann-La Roche Ltd, Pharmaceuticals Division  The most challenging scenario in a Lead Generation project is to target a membrane-bound receptor for which there is no information about small molecule ligands available. In order to tackle this task chemogenomics strategies for GPCR projects have been integrated into the lead generation process at Roche and a substantial investment has been made to compile biological data, to enrich the corporate compound depository with GPCR-specific chemotypes and to provide for a data warehouse to link ligand with target annotations. The application of this strategy within Lead Generation at Roche has successfully identified the first nonpeptidic, small molecule, highly selective somatostatin 5 receptor antagonists.

11:45 Interpreting Compound Profiles in the Context of the Human Drug Experience
	Jacques Migeon, Ph.D., Senior Scientist, BioPrint, CEREP Inc.
Now that compound profiling is ubiquitous, how should these profiles be interpreted? Extensive datasets of in vitro and in vivo information on past and current drugs can be applied to the interpretation of in vitro profiles of new drug candidates. This talk will discuss some of these methods, provide some examples, and provide some insights based on experience with drug profiling.

12:15 pm Panel Discussion: The Value of Doing Systems Biology after HTS Moderator: Mark Murcko, Ph.D. Additional Panelist: Berta Strulovici, Ph.D., Research Vice President, Automated Biotechnology, Merck Research Laboratories  How to use model systems coupled with multiparametric assays Do systems biology assay data extrapolate well to humans?  What is the best order to do assays, parallel vs. sequential?  How predictive are the results?  What have people learned?

12:45 Luncheon Workshop (Sponsorship Available) or Lunch on Your Own

2:00 pm Chairperson’s Remarks
Tudor Oprea, Ph.D.

2:05 Analysis of Biological Mechanism of Action, and Characterization of Profiles across the Progression Path
Dana E. Vanderwall, Ph.D., U.S. Cheminformatics Analysis, MDR Ix Cheminformatics Science & Analysis, GlaxoSmithKline 
>7K compounds were profiled through >460 assays including both in vitro target assays and cell-based phenotypic and pathway assays. These assays represent targets in multiple areas including nuclear receptors, 7-transmembrane receptors, phosphodiesterases, other non-proteolytic enzymes, ion channels, integrins, kinases, proteases, transporters, and 61 primary cell-based phenotypic and biomarker readouts. A suite of statistical methods we're investigated to infer biological mechanisms underlying cellular phenotypes. To investigate whether compound profiling data can be used to understand problems with compound attrition, changes in compound profiles during throughout the progression path were compared, including marketed drugs included as benchmarks. Finally, the relationships within and between protein target classes were investigated, including an analysis of the extent to which compounds synthesized for a target in one class tend to have of “leakage” outside of their target class. 

2:35 Analysis of Drug Induced Effect Patterns for Investigating Systems Structure Effect Relationships
	William T. Loging, Ph.D., Principal Scientist, Medicinal Sciences, Pfizer Global R&D Groton Labs
Molecular medicine is rapidly shifting its focus from the genome towards the characterization of network structures regulating the information flow within the cell and the organism. While methods have been developed for identifying network structures involved in the generation of disease, commensurate approaches for ascertaining network topologies connecting drug effects have not yet become available. Herein we describe a strategy for investigating effects of medicines by determining drug-induced network-topology perturbations. Organism effect information has been linked to medicines inducing these topologies, suggesting that this platform is not only useful for providing insight into system functions, but is also valuable for identifying likely cellular origins of intended and unintended drug effects.

3:05 Characterizing the Multiple Activities of Kinase Inhibitors
	Petra Ross-Macdonald, Ph.D., Senior Research Investigator, Applied Genomics, Bristol-Myers Squibb Co.
Using a novel approach to identify genes that exhibit dose-response behavior, we can examine compound potency, selectivity, on- and off- target activities, and underlying biology across the therapeutic dose range. This strategy is being employed to identify distinct activities within and between compounds.

3:35 Poster Session, Exhibit Viewing, and Refreshment Break

4:00 Compound Profiling by Predicting On-Target Activities, Compound Promiscuity and Adverse Drug Reactions
	Andreas Bender, Ph.D., Lead Discovery Informatics, Center for Proteomic Chemistry, Novartis Institutes for BioMedical Research, Inc.

4:30 Molecular Profiling Driven Drug Discovery: Use of Compound Profiling from Discovery to the Clinic and Back Again
	Stephen Horrigan, Ph.D., Vice President of Research, Avalon Pharmaceuticals
Molecular profiling using genomic approaches is becoming an important process in drug development and is becoming a new paradigm for drug discovery based on complex biomarker sets, or molecular profiles, and is more frequently becoming a standard part of the information used to decide which compounds are advanced in to the clinic and through clinical development. Examples on the use of compound profiling at both early and late stages of the drug-development process through the use of molecular profiles to identify compounds from high throughput screens, reveal their mechanism of action, direct optimization efforts, and identify pharmacodynamic markers for clinical development. This approach leads to an understanding of the biological effects of compounds on a much broader and more fundamental level before they ever get into human clinical trials.

5:00 Panel Discussion: Comparing Available Databases, Public vs. Private

5:30 Close of Conference