Srijit Seal

Principal Scientist at Human Chemical Company and Visiting Researcher at Uppsala University.

I am a researcher in chemoinformatics and computational biology, centered on using machine learning and Cell Painting to study small-molecule bioactivity, image-based profiling, and drug discovery.
I am Principal Scientist at Human Chemical Company and Visiting Researcher at Uppsala University. Previously, I was Senior Scientist at Merck US and 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. I also serve on the Board of Directors at the American Society for Cellular and Computational Toxicology and the Editorial Board of the Journal of Cheminformatics.

AI for Science Cell Painting Bioactivity Modeling Cheminformatics Drug Discovery

30 Publications 856 Citations 151 GitHub stars Updated Oct 4, 2026

Mar 2026

New papers are out in Drug Discovery Today, Cell Painting in primary human hepatocytes, and progress and challenges in image-based profiling.

Feb 2026

Our paper Counting cells can accurately predict small-molecule bioactivity benchmarks is published in Nature Communications.

Sep 2025

PKSmart is published in the Journal of Cheminformatics, and our transfer learning study for ESKAPE antibacterials is published in Chemical Science.

Jan 2025

pip install infoalign! Add biological information to your chemical fingerprints, using only SMILES as input! Our paper Learning Molecular Representation in a Cell has been accepted to ICLR 2025! We introduce InfoAlign, a new approach for learning molecular representations from cellular response data, integrating features like cell morphology and gene expression. By combining information bottleneck methods with context graphs, we’re able to extract minimal yet sufficient representations of molecules that lead to better predictions and generalization in downstream tasks like molecular property prediction and zero-shot molecule-morphology matching.

Mar 2024

Calibrated prediction of scarce adverse drug reaction labels with conditional neural processes has been accepted in DMLR at ICLR 2024. This work introduces a meta-learning approach with neural processes to significantly enhance the accuracy and calibration of adverse drug reaction (ADR) classification.

Srijit Seal

Featured Research

Counting cells can accurately predict small-molecule bioactivity benchmarks

Cell Painting for cytotoxicity and mode-of-action analysis in primary human hepatocytes

PKSmart: an open-source computational model to predict intravenous pharmacokinetics of small molecules

Machine Learning for Toxicity Prediction Using Chemical Structures: Pillars for Success in the Real World

Talks & Workshops

Tools & Tutorials

PKSmart

Introduction to Cheminformatics and AI in Drug Discovery

Leveraging Cell Painting Morphological Profiles for Machine Learning–Driven Bioactivity Prediction

Drug Discovery and Toxicology Utilities

Recent Publications

Model Validation Protocols for Machine Learning in Small Molecule Drug Discovery

Artificial intelligence in drug discovery: what it is, where we stand and the path forward

Human-supervised Agentic AI for Hypothesis Generation and Experimental Assistance in Drug Repurposing

SHOT-CCR: Biologically guided adversarial training for test-time adaptation in cellular morphology

CAPRICHO: Interpretable Quality Flagging and Flexible ChEMBL Bioactivity Curation for QSAR Modeling

Application of machine learning and artificial intelligence methods in predictions of absorption, distribution, metabolism, and excretion properties of chemicals

The OASIS Consortium: integrating multi-omics technologies to transform chemical safety assessment

The medicinal chemist's map to deep learning: Concepts, applications, and case studies

Graph neural processes for molecules: an evaluation on docking scores and strategies to improve generalization

Step Forward Cross Validation for Bioactivity Prediction: Out of Distribution Validation in Drug Discovery

AI agents in drug discovery: applications and case studies

Transfer learning enables discovery of sub-micromolar antibacterials for ESKAPE pathogens from ultra-large chemical spaces

Learning Molecular Representation in a Cell

Progress and new challenges in image-based profiling

Machine Learning for Toxicity Prediction Using Chemical Structures: Pillars for Success in the Real World

Counting cells can accurately predict small-molecule bioactivity benchmarks

Cell Painting for cytotoxicity and mode-of-action analysis in primary human hepatocytes

Advice for Bad Computational Toxicologists

Cell Painting: a decade of discovery and innovation in cellular imaging

Insights into Drug Cardiotoxicity from Biological and Chemical Data: The First Public Classifiers for FDA Drug-Induced Cardiotoxicity Rank

Integrating Cell Morphology with Gene Expression and Chemical Structure to aid Mitochondrial Toxicity detection

Comparison of Cellular Morphological descriptors and Molecular Fingerprints for the prediction of Cytotoxicity- and Proliferation-related assays

Understanding biology with machine learning: compression, intelligibility, and dependency

Using Chemical and Biological data to Predict Drug Toxicity

Merging Bioactivity Predictions from Cell Morphology and Chemical Fingerprint models using Similarity to Training data

From pixels to phenotypes: Integrating image-based profiling with cell health data as BioMorph features improves interpretability

Using Generative Modeling to Endow with Potency Initially Inert Compounds with Good Bioavailability and Low Toxicity

Improved Detection of Drug-Induced Liver Injury by Integrating Predicted In Vivo and In Vitro Data

PKSmart: an open-source computational model to predict intravenous pharmacokinetics of small molecules

Calibrated prediction of scarce adverse drug reaction labels with conditional neural processes

News