Arc Institute's STACK Model Lets Researchers 'Prompt' Cells Like LLMs, Opens Virtual Drug Discovery Atlas
A new foundation model from the Arc Institute treats groups of cells as context windows, enabling zero-shot prediction of drug effects across tissues — without running a single wet-lab experiment.
The Arc Institute has published STACK, a biology foundation model that applies the same contextual reasoning principles behind LLMs to single-cell transcriptomics — and the results are striking enough to warrant attention well beyond computational biology circles. As @Dr_Singularity put it, "Cells can now be 'prompted' like LLMs," and while that's a simplification, it captures the conceptual breakthrough: STACK learns representations from groups of cells that allow researchers to predict the effects of chemical perturbations without ever running the experiment.
Unlock the full briefing
Get every story in today's briefing, the full archive, and the daily AI intelligence brief.
All stories today
Full archive
Daily brief
Cancel anytime. Payments powered by Stripe.