FRIDAY, JUNE 10
8:10 am Chairperson’s Opening Remarks
Jonathan Essex, Ph.D., Head, Computational Systems Chemistry; Chairman, Institute for Complex Systems Simulation (ICSS), School of Chemistry, University of Southampton
8:15 Molecular Dynamics Drug Docking: Modeling Ligand Interactions in the Age of High Performance Computing
Michael Kuiper, Ph.D., Computational Scientist, Victorian Partnership for Advanced Computing
Continuing advances in computational performance now allow researchers to routinely simulate protein molecules in the order of hundreds of nanoseconds. At this timescale it is possible to investigate detailed interactions of ligands with receptors starting with the free ligand in solution. Though not yet suitable for high-throughput drug screening, molecular dynamics drug docking (MDDD) however does offer researchers an approach to observe complex drug/ligand interactions not typically considered in drug design.
8:40 The Internet is Here to Stay: Web Service Delivery of Computational Properties
David Thompson, Ph.D., Senior Principal Systems Engineer, Boehringer Ingelheim GmbH
A robust and extensible web services framework for the delivery of computational properties to the medicinal chemist’s desktop will be presented. This architecture fully leverages our High Performance Compute environment, exposes a wide variety of computational engines, and can be utilized in a manner that best fits the scientists’ requirements. Use cases including the consumption of in silico physicochemical properties and the distribution of quantitative structure-activity relationship (QSAR) models will be described.
9:05 The Binding Energy Distribution Analysis Method (BEDAM) for Structure-Based Drug Design: Theory and Applications
Ronald M. Levy, Ph.D., Board of Governors, Professor of Chemistry & Chemical Biology, Rutgers University
The binding energy distribution analysis method (BDEAM) for structure-based drug design is a new approach to computing protein-ligand binding free energies which makes use of replica exchange molecular dynamics simulations to compute absolute binding affinities. The balance between binding enthalpy and entropy is seen in our formalism as a balance between unfavorable and favorable binding modes. Both the theory and application of BEDAM will be discussed.
9:30 Fragment-Based Screening by Free Energy Simulations
Jonathan Essex, Ph.D., Head, Computational Systems Chemistry; Chairman, Institute for Complex Systems Simulation (ICSS), School of Chemistry, University of Southampton
Free energy simulation is potentially a very powerful tool for structure-based drug discovery. In this presentation, the application of a number of these techniques to locate and score molecular fragments and water in protein binding sites are described. These approaches offer advantages over more conventional simulation methods in that not only is fragment binding ranked in terms of free energy (i.e. entropy is included), but all fragments are in direct competition with water for the binding site.
9:55 A New Computational Method for Predicting Binding Free Energies of Protein-Ligand Interactions
Christopher Langmead, Ph.D., Associate Professor of Computer Science, Carnegie Mellon University
This presentation will discuss a new computational method, called GOBLIN, for performing physics-based free energy calculations under protein and ligand flexibility. GOBLIN compactly encodes Boltzmann distributions over structures by exploiting conditional independencies. Results on HIV-1 PR will be presented demonstrating that it achieves superior quantitative accuracy than competing methods.
Selected Poster Presentation:
10:20 Small Covalent Peptidomimteic Inhibitors of Crm1 Mediated Nuclear Transport
Sharon Shechter, Head of Computational Discovery, Karyopharm Therapeutics Inc.
Nucleo-cytoplasmic transport of macromolecules is a fundamental process of eukaryotic cells. Translocation of proteins and many RNAs between the nucleus and the cytoplasm is carried out by shuttling import and export receptors. CRM1 (Xpo1) is a major exporter for proteins from the nucleus to the cytoplasm, including tumor suppressors (TSPs) and other growth regulatory proteins (GRPs) such as p53, FOXO, pRB, p21, p27, BRAC1/2 and IkB. Here, we describe the identification of novel Crm1 inhibitors using a hierarchical structure-based discovery process which evolved along with the accumulation of novel experimental data providing new insights about CRM1 structure and function. A computer assisted method for the generation of ‘hyper-receptor’ models of Crm1, based on the x-ray structure of CRM1 was used. A ‘hyper-receptor’ model is a model that incorporates additional conformational adjustments required by different types of scaffold. The model built used three-dimensional alignment of known of Crm1 inhibitor originated in published data or generated in house. As a result, the putative ligand binding site was constructed and key residues were highlighted and used for virtual screening. This model could rationalize LMB’s and other published compounds activity. Our screening involved focused libraries designed to contain compounds possessing fragments with a potential to behave as covalent inhibitor (either reversible or irreversible). This has led to the discovery of a series of inhibitors which were more potent and more diverse than the one published.
10:35 Networking Coffee Break in the Exhibit Hall with Poster Viewing
» KEYNOTE PRESENTATION
11:00 Millisecond-Long Molecular Dynamics Simulations of Proteins on a Special-Purpose Machine
 David E. Shaw, Ph.D., Chief Scientist, D. E. Shaw Research and Senior Research Fellow, Center for Computational Biology and Bioinformatics, Columbia University
Molecular dynamics simulation provides a potentially powerful tool for understanding the behavior of proteins at an atomic level of detail, but its relevance to drug design has previously been limited in part by the computational demands of such simulations. We have constructed a specialized supercomputer, called Anton, that has simulated the behavior of a number of proteins for periods as long as a millisecond -- approximately 100 times the length of the longest such simulation previously published -- revealing pharmaceutically relevant aspects of protein dynamics that were previously inaccessible to both computational and experimental study.
|
Sponsored by
11:45 SZMAP: Mapping Solvent Thermodynamics in Binding SitesAnthony Nicholls, Ph.D., President & CEO, OpenEye Scientific Software
Semi-continuum solvent theory captures discrete effects that can be important in enclosed spaces such as binding cavities. Using this model, SZMAP rapidly maps thermodynamic quantities of water molecules near protein surfaces by employing a single explicit water probe. The resulting quantities may be used as a correction factor for continuum solvent calculations as well as serving to guide the design of ligand analogues and optimizing binding affinity.
Sponsored by.jpg "Intelligent Pharma")
12:05 pm Molecular Field Based Virtual Screening and Molecular Design
Jascha Blobel, Ph.D., Product Manager, Sales & Product Development, Intelligent Pharma
By comparing molecular fields amongst molecules, it is possible to find structurally different molecules with the same biological functions. The molecule mimics are generally selected from a compound database. Depending on the database type, different information can be deduced, such as finding mechanisms of action, hit identification, etc. However, the use of predefined databases sets limitations in the identification of innovative molecules. In order to overcome this problem, Intelligent Pharma has developed an artificial intelligence system with a novel molecule designer which finds innovative active molecules in the unexplored chemical space.
12:20 Luncheon Presentations (Sponsorship Opportunities Available) or Lunch on Your Own
1:30 Chairperson’s Remarks
Ruben Abagyan, Ph.D., Professor, Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California, San Diego
1:35 Structure-Based Ligand Discovery for GPCRs
Ruben Abagyan, Ph.D., Professor, Skaggs School of Pharmacy and Pharmaceutical Sciences, University of California, San Diego
The recent structures of GPCRs give us a better understanding of the binding pockets for both antagonist and agonists and insights into the structural mechanism of the receptor activation. Computational approaches to structure-based docking and modeling of pockets of subtypes and homologues are presented along with successful application of these methods to finding and optimizing GPCR modulators. Results of the recent docking and modeling assessment (a.k.a. GPCR Dock 2010) for CXCR4 with small molecule and a peptide, as well as the dopamine receptor D3, are reviewed.
2:00 The Role of Recent Crystal Structures of Membrane Bound Proteins in Drug Discovery for CNS Targets
Sid Topiol, Ph.D., CSO, Computational and Structural Investigations, 3D-2Drug
I would present illustrations of the impact of recent X-ray structure determinations for the more challenging CNS targets, i.e., membrane bound proteins. For C GPCR’s, new opportunities for drug discovery are being identified using X-ray structures of the extracellular regions. Other targets, such as transporters and ion channels, are also now amenable to structure-based drug design.
2:25 Hitting a Moving Target: Characterizing GPCR Signaling through Long-timescale Molecular Dynamics Simulations
Ron Dror, Ph.D., Senior Research Scientist and Special Advisor to the Chairman, D. E. Shaw Research
A mounting body of evidence indicates that drugs induce GPCRs to interconvert between numerous conformational states with distinct intracellular signaling profiles. Recent advances in algorithms and hardware for molecular dynamics (MD) simulations are now bringing the previously inaccessible timescales on which these transitions occur within reach. This talk will describe ongoing studies of GPCRs using state-of-the-art MD simulations, which have provided a hitherto elusive glimpse of the conformational dynamics underlying GPCR-mediated signaling by both endogenous ligands and drugs.
2:50 Networking Refreshment Break
3:00 Beyond the Orthosteric Binding Site: A Structure-Based SAR Analysis of the D3R Selective Compounds
Lei Shi, Ph.D., Assistant Professor, Department of Physiology and Biophysics, HRH Prince Alwaleed Bin Talal Bin Abdulaziz Alsaud Institute for Computational Biomedicine, Weill Cornell Medical College
Selective targeting of dopamine D3 receptor (D3R) has therapeutic implications in neuropsychiatric disorders and drug additions. D3R selective compounds have two pharmacophores and a connecting linker. The talk will highlight individual and combined contributions of these components towards the selectivity in the context of the D3R structure. A novel structure-based design scheme to address specificity issues of highly homologous GPCRs will be presented.
3:25 Moving in New Circles – Exploiting Macrocycles for Drug Discovery
Nick Terrett, Ph.D., CSO, Ensemble Therapeutics Corp
Macrocycles are largely underexploited in drug discovery because they are generally perceived as structurally complex and difficult to access. Ensemble Therapeutics has developed platforms for the rapid synthesis and screening of macrocycles in order to identify leads for challenging protein-protein interaction targets. The talk will focus on the design and synthesis of macrocycle libraries and the successful discovery of novel lead molecules with unprecedented activity and drug-like properties.
3:50 Strategizing to Develop Resistance-Proof Inhibitors
Bruce Tidor, Ph.D., Professor of Biological Engineering and Computational Science, Massachusetts Institute of Technology
The selection of resistant variants is an important problem limiting the therapeutic usefulness of inhibitors to targets undergoing rapid mutation, particularly for applications in infectious disease and cancer. We report our work exploring general strategies for the development of inhibitors that have a reduced tendency to induce resistance, using HIV protease as a trial target.
4:15 Computational Approaches to Modeling the Emergence of Drug Resistance
Ryan Lilien, M.D., Ph.D., Assistant Professor, Department of Computer Science & Donnelly Centre for Cellular and Biomolecular Research, University of Toronto
The emergence of drug resistance reduces the effectiveness of many novel therapeutics. I will describe computational methods for predicting resistance mutations through their structural and functional effects on the protein target. These methods may allow us to identify new ways to create drugs that are less likely to be made ineffective by pathogen evolution, understand the key determinants of the evolution and spread of resistance, and develop the ability to slow the emergence of resistant variants.
4:40 End of Conference