Virtual-cell models compress unseen intervention geometry through a target-specific generalization bottleneck

Abstract excerpt

Predictive models of cellular perturbation are often judged by how closely they reconstruct molecular states after unseen interventions. We show that high state-level similarity can coexist with loss of the relationships that distinguish perturbations, a failure we term Intervention Geometry Compression (IGC). Across established models and perturbation settings, unseen interventions show weakened global and local...

Topics

Open a Topic to create a Post that cites this publication.

• Artificial Intelligence
• Computer Simulation
• Deep Learning
• Drug Evaluation, Preclinical
• Phenotype

Identifiers and source

Literature Corpus work

b4c84d18-ef8d-5561-a418-bf939c32f040

DOI

10.64898/2026.08.21.746243