Srijit Seal

Short presentation

I am a researcher in chemoinformatics, centered on using machine learning techniques, particularly modeling, and interpretation of the Cell Painting assay, to predict drug bioactivity, safety, and toxicity.

Previously, I was a Senior Scientist at Merck US. I completed my postdoc at the Broad Institute of MIT and Harvard where I was advised by Anne Carpenter and Shantanu Singh.

I obtained my PhD from the University of Cambridge where I was advised by Andreas Bender.

Keywords

• Cell Painting
• Machine Learning
• Drug Discovery
• Chemoinformatics
• Toxicity Prediction
• Pharmacokinetics
• Bioactivity
• Image-Based Profiling
• Mechanism of Action
• DILI
• PKSmart
• High-Content Screening
• Generative Modeling
• Multi-omics Integration
• AI Toxicology
• PhenQSAR
• Morphological Profiling
• Small Molecules
• Drug Repurposing
• Cell Health

Biography

Srijit Seal is a prominent computational scientist specializing in the intersection of artificial intelligence, chemoinformatics, and drug discovery. Prior to his current role, Seal served as a Senior AI Scientist at Merck and maintains research affiliations with the University of Cambridge and the Broad Institute of MIT and Harvard. His academic foundation was built at St. Stephen’s College in Delhi before he moved to the University of Cambridge to complete his PhD under the guidance of Andreas Bender. During his doctoral studies, he focused heavily on leveraging machine learning to bridge the gap between chemical structures and biological responses, establishing himself as a leader in predictive toxicology.

Research

His research is primarily defined by the innovative use of image-based profiling, specifically the Cell Painting assay, to predict drug safety and efficacy. By analyzing high-dimensional morphological data from cells, Seal has developed methods to identify potential side effects like drug-induced liver injury and cardiotoxicity much earlier in the development process. One of his most significant collaborative achievements involved leading a large-scale study with over 30 scientists from major pharmaceutical companies like Pfizer and AstraZeneca to define the "Pillars for Success" for AI-driven toxicology. This work helped standardize how the industry validates and uses machine learning models for regulatory safety assessments.

Beyond theoretical research, Seal is well-known for creating accessible, open-source computational tools that benefit the broader scientific community. He developed PKSmart, a platform designed to predict human pharmacokinetic parameters, and DILIPredictor, which focuses on early-stage liver toxicity detection. His work often emphasizes "BioMorph" features, which aim to make complex machine learning outputs more interpretable for biologists by linking mathematical patterns to recognizable cellular health indicators. These contributions have earned him numerous accolades, including multiple awards from the Society for Laboratory Automation and Screening (SLAS) and the prestigious SLAS Authors’ Choice Award.

Seal is also a dedicated leader within the global scientific community, serving on the Board of Directors for the American Society for Cellular and Computational Toxicology. He frequently shares his expertise through international seminars, particularly focusing on the ethical and practical implementation of AI in pharmaceutical development. As an editorial board member for the Journal of Cheminformatics, he continues to influence the direction of the field by advocating for robust, reproducible, and biologically grounded computational models.

Media

Dr Srijit Seal leads study to define AI-powered toxicology guidelines

The review proposes a five-pillar framework for reliable, actionable toxicity modeling, and draws from the expertise of scientists from companies including GSK (GlaxoSmithKline), Novartis, Eli Lilly..

https://www.clarehall.cam.ac.uk/news/dr-srijit-seal-leads-study-to-define-ai-powered-toxicology-guidelines/