Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Transcriptomics
Computational methods resolve cell and spatial structure in transcriptomic data
Recent work pairs new spatial and single-cell algorithms with cross-method benchmarks and reproducible pipelines, tightening evaluation of computational…
ML / AI methods
Pretrained protein and RNA language model embeddings feed diverse prediction tasks
Frozen and fine-tuned sequence model embeddings now extend to stability, PTM, binding-site, and structure-aware prediction tasks.
Genomics
Causal and polygenic frameworks extend genomic risk prediction and mediation analysis
Scalable causal and risk-prediction methods now address heterogeneity, pleiotropy, and multi-ancestry transfer in omics and genetic data.
ML / AI methods
Interpretable ML and foundation models extend single-cell and genomic analysis
Foundation-model embeddings are being paired with interpretability and benchmarking requirements across single-cell, variant, and drug-response tasks.
Structural bio & protein design
Explicit structural geometry improves antibody binding and epitope prediction
Recent preprints and journal articles pair explicit structural geometry with sequence language models for antibody, peptide and epitope prediction tasks.