Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Single-cell
Histology-to-transcriptomics models now target programs, priors, and shared latent spaces
Preprints now reformulate histology-to-spatial-transcriptomics prediction using transcriptional programs, pathway priors, and disentangled representations.
Structural bio & protein design
Computational models predict peptide-HLA binding, antibody structure similarity, and…
Learned and structure-based representations now drive binding, clustering, and vaccine-design predictions across immunology and antiviral targets.
ML / AI methods
Structure-aware and sequence-based encoders learn reusable RNA and DNA representations
Recent August 2026 reports extend pretrained sequence models to RNA binding profiles and structure-guided m6A prediction.
Single-cell
Single-cell profiles link cell states to disease, mortality and drug combinations
Single-cell transcriptomes are now anchored to plasma proteomes, tissue damage axes and drug-combination design across human tissues.
ML / AI methods
Pretrained protein models drive search, alignment, and ensemble generation
Recent preprints and a ProteinDPO publication couple frozen or fine-tuned protein models to tree search, preference alignment, and ensemble generation.