Click to perturb — we pioneer generative AI models for in silico perturbation of cells, predicting how they transition between states in response to drugs, gene knockouts, and other interventions. Learn more →

Lotfollahi Lab

Machine Learning ∩ Genomics ∩ Biomedicine

Wellcome Sanger Institute & University of Cambridge

Research Directions

Perturbation Modeling

Generative AI to predict cellular responses to drugs, CRISPR, and disease — enabling counterfactual biology and virtual screening.

→

Spatial Genomics

Graph deep learning for tissue architecture, niche discovery, and in silico perturbation of tissue ecosystems.

→

Tissue Foundation Models

Multimodal generative models linking histology, spatial transcriptomics, and clinical data for 3D tissue understanding.

→

Core ML Research

New architectures, training objectives, and probabilistic methods for scalable biomedical generative modeling.

→