DynaVelo model predicts B-cell perturbation effects from single-timepoint multiome data
Start here
A study published in Nature Biotechnology introduces DynaVelo, a generative deep learning model that uses a neural ordinary differential equation to jointly model RNA expression and transcription factor motif accessibility. Working from single-time-point paired single-cell RNA and ATAC multiome data, it infers cell dynamics, predicts loss-of-function perturbation outcomes in B cells, and nominates rescue targets. The work points toward computational screening of perturbations and rescues without time-course experiments. What remains open is how the approach performs beyond B cells, in other tissues and disease settings, and whether nominated rescue targets hold up experimentally in vivo.
Written by a model from the reports · updated 3 hours ago
Coverage timeline
Follow the coverage to see the different angles.
- NaturePickDeep dynamical models of single-cell multiomic velocities predict loss-of-function and rescue perturbations in B cells
The authors present DynaVelo, a generative deep learning model that uses a neural ODE to jointly model RNA expression and TF motif accessibility. It infers cell dynamics from single-time-point paired scRNA and scATAC multiome data, predicts loss-of-function perturbation outcomes, and nominates rescue targets.
Coverage and heat for this story
Reports so far 1·Heat now 7·Peak heat 10 (Oct 21 08:00)·Reports, 24h +1
Two panels, same clock (pointer stays in sync): top is cumulative editorial reports — only rises; last point = All reports; slope = new reports that hour. Bottom is Live-board heatIndex — 7-day participants with a 24-hour half-life, so it can fall when talk cools.