The 19th Annual Symposium of the Institute of Chemical Biology and Drug Discovery (ICB&DD), “Drug Discovery & AI: Advances and New Directions,” attracted a broad audience and provided an opportunity to learn about and discuss the current state and the future of drug discovery and development in light of the breathtaking advances in artificial intelligence (AI).
The symposium shed light on the impact that AI has started to have on drug discovery and the significance of integrating AI methods with fundamental theory and methods of molecular biology and physical sciences to generate interpretable AI-driven discoveries in pharmacology.
The symposium, held at the Charles B. Wang Center, was organized by event chair Ivet Bahar, director of the Laufer Center for Physical and Quantitative Biology, and co-chair Dima Kozakov, professor of Applied Mathematical Sciences and Laufer Center faculty (currently at Oden Institute for Computational Engineering and Sciences, UT Austin). It was sponsored by the Office of the Vice President for Research, the Renaissance School of Medicine, the College of Arts and Sciences Department of Chemistry, Hoffman and Baron LLP, and Chembio Diagnostics Systems Inc.
In the opening session, Bahar welcomed attendees and gave a brief history of the importance of artificial intelligence in modern science. Stony Brook Executive Vice President and Provost Carl Lejuez gave opening remarks, which highlighted the importance of multidisciplinary entities on campus such as ICB&DD and its long-lasting record of attracting some of the best faculty to the SBU scientific community. Iwao Ojima, director of the ICB&DD and distinguished professor in the Department of Chemistry, followed with an overview of the institute, its history and mission.
AI has impacted various areas of biology and biomedicine, including structural biology and personalized medicine, and its impact on drug discovery is yet to be established. Yet, several talks by distinguished speakers highlighted notable accomplishments enabled by novel AI methods, signaling that AI has already begun to accelerate computer-aided drug discovery and facilitate the design of novel therapeutic interventions.
Seven distinguished speakers from academia and industry, and one invited speaker from industry, delivered insightful talks throughout the symposium:
Chris Sander, Harvard Medical School, presented how the tools developed in his lab are revolutionizing molecular biology and cancer biology research and assisting in designing combinations of targeted compounds rather than general chemotherapeutics. Two such tools are EVcouplings, based on the fundamental concept of linking sequence coevolutionary relations to structural connectivity, and CancerRiskNet for identifying high risk of cancer from clinical records using machine learning.
Nikolay Dokyolyan, University of Virginia School of Medicine, emphasized the importance of merging physics-based modeling with AI, and presented a newly developed AI-guided platform, YuelDesign, for de novo drug design.
Marta Filizola, Icahn School of Medicine at Mount Sinai, NY, showed the power of AI coupled with high resolution GPCR structures and ultra-large libraries of small molecules for refining GPCR-targeted therapeutics.
David Koes, University of Pittsburgh, developer of widely used tools such as Pharmit and Gnina, presented recent advances in deep learning methods for efficient docking and scoring of ligands in virtual screening campaigns.
Maria Wendt, global head of Preclinical Computational Innovation Strategy for R&D at Sanofi in Cambridge, MA, described the Biologics x AI Moonshot program at Sanofi, which uses Protein Language Models and deep learning approaches.
Patrick Walters, chief scientist at OpenADMET in Cambridge, MA, showed how active learning, a machine learning technique, can help in efficiently evaluating relative binding free energies of drugs onto proteins.
Dima Kozakov, co-developer of ClusPro for protein docking and FTMap for hotspot identification, presented recent advances in bridging atomic scale modeling and proteome-scale interaction networks to uncover actionable targets for therapeutic intervention.
David LeBard, OpenEye and Cadence Molecular Sciences, NM, described a recently developed workflow that detects cryptic pockets for drug binding.
This year’s symposium poster session, chaired by Anupam Banerjee, featured 83 posters reflecting the range and vitality of research in medicinal chemistry, structural biology, computational biology and related areas. The session showcased the depth and diversity of research at Stony Brook and highlighted the creativity, collaboration and scientific excellence of our research activities that contribute to drug discovery.
The event drew strong participation from students, postdoctoral researchers and faculty, who shared their findings and exchanged ideas in a lively and engaging setting. Six judges — Eugene Serebryany, Dale Drueckhammer, Jeffrey L. Gustafson, Jessica Seeliger, Wenchao Qu and Maurizio Del Poeta — evaluated the posters for scientific merit, creativity and clarity.
Awards were presented in three categories:
In Computational Biology, Xiaowei Bogetti (postdoctoral associate, Laufer Center) was recognized for “smFRET/DEER-guided Modeling of Protein Conformational Dynamics for Rational Drug Design” (co-authored by Ivet Bahar).
In Medicinal Chemistry, Anushka Ojha (PhD student, Molecular and Cellular Biology Program, Department of Pathology) received the award for “Beyond 5-FU: Exploring the Efficacy of Gemcitabine-modified miRNA-15a in Resistant CRC” (co-authors: Amartya Pal and Jingfang Ju).
In Structural Biology, Archana Sudevan (PhD student, Department of Chemistry) was honored for “NO Driven Dual Regulation of Nitric Oxide Sensing Protein (NosP) Through Interdependent Cysteine S-Nitrosylation and Heme Binding in Burkholderia Thailandensis,” co-authored by Sweta Anantharaman, Jiayuan Fu and Elizabeth M. Boon.
Add comment