Inferring stochastic dynamics by biophysical Neural ODE using single-cell transcriptomics
Chunhe Li
Nature Communications, Published online: 19 May 2026; doi:10.1038/s41467-026-73257-z Dou and colleagues present DynNet, a biophysically informed deep learning framework that reconstructs continuous cellular dynamics from time-resolved single-cell data. DynNet maps cell fate transitions and offers insights into complex biological processes.