Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Structural bio & protein design
Structure-based methods predict ligand binding sites, poses and cryptic pockets
Co-folding and geometric learning models now target apo, cryptic, and allosteric sites while new benchmarks test co-folding on hundreds of ligand complexes.
Single-cell
Computational methods resolve tumor clonal structure from single-cell and long-read data
Low-coverage single-cell, nanopore, and methylation datasets demand scalable tools that link somatic variation to clonal and epigenetic states.
ML / AI methods
Generative and predictive models span DNA, enzyme, metabolic and single-cell data
Unified benchmarks and topology- and alignment-aware encodings are now appearing across enzymatic, genomic and single-cell modeling.
Single-cell
Computational methods extract cell-level structure from single-cell and spatial data
Growing single-cell and spatial datasets demand methods that test analytical stability, infer cell-cell communication, and resolve spatial dynamics.
Transcriptomics
Bias-aware and multi-dataset models refine ORF calls from ribosome profiling data
Protocol biases and read-mapping choices affect ORF calls, prompting bias-aware, multi-dataset, and configurable ribosome profiling methods.