The second joint showcase between NCI and Intersect Australia is entitled:
AI and ML in Drug Design and Discovery
Topics and Plenary Speakers:
Prof Olexandr Isayev (Carnegie Mellon University)
Prof. Olexandr Isayev is an Assistant Professor at the Department of Chemistry at Carnegie Mellon University. In 2008, Olexandr received his Ph.D. in computational chemistry. He was Postdoctoral Research Fellow at the Case Western Reserve University and a scientist at the government research lab. During 2016-2019 he was a faculty at UNC Eshelman School of Pharmacy, the University of North Carolina at Chapel Hill. Olexandr received the “Emerging Technology Award” from the American Chemical Society (ACS) and the GPU computing award from NVIDIA. The research in his lab focuses on connecting artificial intelligence (AI) with chemical sciences.
Accelerating Drug Discovery with Machine Learning and AI
Deep learning is revolutionizing many areas of science and technology, particularly in natural language processing, speech recognition and computer vision. In this talk, we will provide an overview into latest developments of machine learning and AI methods and application to the problem of drug discovery and development at Isayev’s Lab at CMU. We identify several areas where existing methods have the potential to accelerate pharmaceutical research and disrupt more traditional approaches.
First we will present a deep learning model that approximate solution of Schrodinger equation. Focusing on parametrization for drug-like organic molecules and proteins, we have developed a single ‘universal’ model which is highly accurate compared to reference quantum mechanical calculations at speeds 10^6 faster.
Second, we proposed a novel computational method for de-novo design of molecular compounds with desired biological properties. General workflow of the proposed method integrates two deep neural networks – generative and predictive – that are initially trained separately but then trained jointly to generate novel chemical structures with desired properties. In the proof-of-concept study, we have employed this integrative strategy to design chemical libraries biased toward compounds with either maximal, minimal, or specific ranges of physical properties (like solubility and hydrophobicity) as well as to develop novel kinase inhibitors of JAK2, EGFR, CDK1, and TBK1. This new approach can find general use for generating targeted chemical libraries optimized for a single desired property or polypharmacology.
Dr. Arvind Ramanathan (Argonne National Lab)
Dr. Arvind Ramanathan is a computational biologist in the Data Science and Learning Division at Argonne National Laboratory and a senior scientist at the University of Chicago Consortium for Advanced Science and Engineering (CASE). His research interests are at the intersection of artificial intelligence, high performance computing and biological/biomedical sciences. His research focuses on developing scalable AI methods for understanding complex biological phenomena including phase separation in biological systems as well as design of novel CRISPR/Cas9 probes to modify microbial functions. He obtained his PhD from Carnegie Mellon University and has been awarded the UT-Battelle Early Career Award (2017), apart from being recognized with the IEEE/ACM Gordon Bell Award for HPC in COVID-19 research (2019).
A supercomputing perspective on AI-driven drug discovery for COVID-19
The landscape of supercomputing infrastructure has been continually evolving with the development of novel AI/ML approaches, where parameter rich AI-models are now driving how scientific applications are ’steered’ or in some cases, even ‘replaced’ with models that are computationally more efficient. We outline our experience in developing AI/ML methods in the context of structural biology applications to discover inhibitors against the novel coronavirus disease (COVID-19) causing agent, namely the severe acute respiratory coronavirus 2 (SARS-CoV-2). We demonstrate how AI/ML models can provide realistic speedups for large-scale virtual screening experiments; further, a novel workflow that integrates virtual screening with adaptive sampling molecular dynamics simulations led to drastic reduction in time-to solution for discovering new inhibitors targeting SARS-CoV-2. By integrating experimental data into our workflows, we show that we can obtain novel insights into the mechanism of various SARS-CoV-2 protein targets.
Prof Cecilia Clementi (Freie Universitaet Berlin)
Designing Molecular Models with Machine-Learning and Experimental Data
Prof Michele Vendruscolo (Uni of Cambridge)
(Image: liposome – Credit to A/Prof Defang Ouyang at Uni of Macau)