Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Single-cell
Graph and contrastive models recover spatial and single-cell transcriptomic structure
A May 2026 batch of graph, contrastive, and diffusion methods addresses incomplete spatial and single-cell transcriptomic measurements.
Single-cell
New computational methods benchmark and mine single-cell, spatial and bulk omics data
Recent preprints and papers benchmark spatial platforms and single-cell integration models, with new tools for QC, marker discovery and trajectory analysis.
Structural bio & protein design
Deep models couple sequence generation with structure prediction for proteins, peptides,…
Recent preprints and publications apply co-generative language models, diffusion, and Boltz-2 fine-tuning to protein, peptide, and RNA design.
ML / AI methods
Language model embeddings are expanding protein and gene function prediction methods
Recent papers fuse language model embeddings with graph, contrastive, and diffusion methods for low-homology and data-limited prediction.
Genomics
Deep learning models for variant effect and genome annotation across species
Genome-scale benchmarks now test variant pathogenicity and splice-altering predictors across human and non-human species.