Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Single-cell
Generative flow and diffusion models predict single-cell perturbation responses
Recent preprints and publications apply flow matching and latent diffusion to single-cell perturbation and cross-modal prediction.
ML / AI methods
Inferring spatial expression from H&E histology via generative and retrieval models
Between August 8 and 14, 2026, several H&E-to-expression models appeared, using contrastive, diffusion, and single-cell reference strategies.
ML / AI methods
Deep learning models predict nucleic acid interactions and RNA modification sites
Pretrained RNA and protein language model embeddings are being fused with structural and graph features across RNA, DNA and binding prediction tasks.
Systems biology
Uncertainty-aware and spatially structured statistical models for multi-omics data
Spatial and multi-omics datasets demand estimators that encode spatial, replicate and uncertainty structure rather than assuming independence.
Tools & infrastructure
Open tools and databases for pangenome, plasmid, and gene-history inference
Recent releases provide precomputed pangenome resources and updated tools for orthology, plasmid reconstruction, and ancient DNA metagenomics.