Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Single-cell
Computational frameworks resolve cell states, spatial niches and perturbation effects
Spatial CRISPR screening, single-cell foundation model attribution and new Perturb-seq data expand single-cell and spatial readouts at increasing scale.
Structural bio & protein design
Learned models generate and score biomolecules guided by structural data
Pretrained co-folding, docking, and generative models are increasingly steered by experimental restraints and budget-aware search strategies.
Genomics
Computational frameworks now score missense, splicing and promoter variant effects
Large experimental and foundation-model datasets now allow calibrated benchmarking of variant effect predictors across exons, promoters and proteomes.
Single-cell
Statistical methods integrate chromatin and perturbation data at single-cell resolution
New methods pair Perturb-seq, CUT&Tag, and scATAC-seq data with explicit causal, noise, or power models to resolve regulatory mechanisms.
ML / AI methods
Deep sequence models predict variant and expression effects across genomic tasks
Recent work tests whether generalist LLMs, personal-genome fine-tuning, and per-variant confidence estimation improve sequence-based genomic predictions.