BioML Seminar 3.1 - Modeling cell perturbations with STATE

​​[IN PERSON EVENT IN BERKELEY, LOCATION TBA]

​​​​Join us for a new seminar from the BioML group in Machine Learning at Berkeley, sponsored by Amplify Partners. We're excited to host Abhinav Adduri, Machine Learning Research Scientist at the Arc Institute, who'll be presenting his work on STATE: a transformer model that predicts cell perturbations.

​Talk Details: Cellular responses to perturbations are a cornerstone for understanding biological mechanisms and selecting drug targets. While machine learning models offer tremendous potential for predicting perturbation effects, they currently struggle to generalize to unobserved cellular contexts. Here, we introduce STATE, a transformer model that predicts perturbation effects while accounting for cellular heterogeneity within and across experiments. STATE predicts perturbation effects across sets of cells and is trained using gene expression data from over 100 million perturbed cells. STATE improved discrimination of effects on large datasets by more than 30% and identified differentially expressed genes across genetic, signaling and chemical perturbations with significantly improved accuracy. Using its cell embedding trained on observational data from 167 million cells, STATE identified strong perturbations in novel cellular contexts where no perturbations were observed during training. Overall, the performance and flexibility of STATE sets the stage for scaling the development of virtual cell models.

​Speaker details: Abhinav did his undergraduate at UC Berkeley, majoring in Computer Science and Biology. He completed his Ph.D. in Computational Biology and Machine Learning at Carnegie Mellon University, where he developed generative ML algorithms for analyzing DNA and predicting small molecule and protein interactions in silico. He is now a Machine Learning Research Scientist at the Arc Institute, where he builds foundation models for biology, with an emphasis on perturbation response prediction (e.g., characterizing how cells change in response to drug treatments) and reasoning models.