Л. А. Урошлев

This study evaluates machine learning's ability to predict ancient DNA age from damage profiles, finding no significant predictive power beyond baselines.

AutoRNA uses a variational autoencoder to predict RNA tertiary structures from sequences, achieving reasonable accuracy but limited global fold recovery.

This study uses single-cell transcriptomics to show zebrafish pigment cells develop directly from broadly multipotent progenitors, challenging traditional fate restriction models.