Brief

Researchers have developed CellFluxRL, a novel framework for creating virtual cells that adhere to biological and physical constraints. This approach uses reinforcement learning with biologically meaningful reward functions to improve upon existing generative models. The resulting CellFluxRL model demonstrates enhanced biological function, structural validity, and morphological correctness compared to its predecessor, moving towards more biologically meaningful simulations for applications like drug discovery. AI

Researchers have developed a new framework called Bayesian Stochastic Flow Matching (BSFM) to improve the reliability and accountability of generative models used in scientific imaging. This approach builds upon Stochastic Flow Matching (SFM) by incorporating uncertainty quantification, allowing for better generalization across different experimental conditions and the detection of unreliable predictions. Experiments on cellular imaging and fMRI data demonstrate that BSFM effectively provides anomaly scores for detecting out-of-distribution cases within practical sampling budgets. AI