Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
ML / AI methods
Deep and representation learning models integrate spatial, multiomic, and protein data
Multi-modal generative, contrastive, and graph models are now applied to spatial transcriptomics, multiomics, and protein function data.
Single-cell
Inferring regulatory activity from single-cell, nuclear, and cell-free DNA readouts
Single-cell and cell-free DNA profiling now resolves enhancer activity, splicing states, and chromatin accessibility QTLs across tissues and disease contexts.
ML / AI methods
Computational models map sequence variation to protein-level properties, with validation
Benchmarking against large experimental datasets now quantitatively tests energy-based and language-model predictions of sequence effects.
Tools & infrastructure
Open-source software now covers pathology, omics, and long-read epigenetic analysis
Recent preprints and publications release installable tools for preprocessing, visualization, and modeling of multimodal biological data.
ML / AI methods
Interpretable deep models for DNA sequence-to-function and regulatory prediction
Recent preprints test whether sequence-based deep models for regulatory genomics yield interpretable predictions, not just accurate ones.