Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
ML / AI methods
Benchmarks and new frameworks for spatial omics domain, gene, and cell-type inference
Spatial omics methods now pair uncertainty quantification and cross-modal integration with systematic benchmarking of existing tools.
8 preprints · 12 published · ↑ 3.1× vs 4-wk avg
Rank 2 Systems biology
Network-aware methods refine transcriptomic inference and target prioritization
Routine transcriptomic workflows increasingly demand explicit accounting for network structure and technical variation in both analysis and evaluation.
5 preprints · 8 published · ↑ 2.1× vs 4-wk avg
Rank 3 ML / AI methods
Protein language models gain structural and chemical priors for design and binding tasks
Preprints and peer-reviewed papers now test whether structural and chemical priors improve protein language models for prediction and design.
9 preprints · 3 published · ↑ 2.0× vs 4-wk avg
Rank 4 Single-cell
Deep generative and graph models address single-cell and spatial transcriptomics tasks
Recent work applies hybrid deep architectures to practical bottlenecks in imputation, panel design, batch correction, and gene regulatory inference.
1 preprint · 11 published · ↑ 2.0× vs 4-wk avg
Rank 5 Genomics
Benchmarking and new callers for structural, copy number, and repeat variants
Recent preprints and publications assess variant callers and genotypers against specific error classes across long-read, exome, and liquid biopsy data.
3 preprints · 8 published · ↑ 1.9× vs 4-wk avg
Rank 6 Learned representations of protein interfaces guide TCR, antibody, and peptide design↑ 1.7×Structural bio & protein designRank 7 Graph and transformer models learn protein pockets and ligands for drug discovery↑ 1.7×ML / AI methodsRank 8 Deep learning predicts expression, chromatin loops and TF binding from DNA sequence↑ 1.7×EpigenomicsRank 9 Learned sequence representations underpin protein interaction and activity prediction↑ 1.6×ML / AI methodsRank 10 Computational methods recover cell and spatial organization from spatial transcriptomics↑ 1.6×TranscriptomicsRank 11 Regulatory variants and enhancer maps resolve noncoding control of developmental genesEpigenomicsRank 12 Graph and transformer models predict and generate biomedical molecules and peptidesML / AI methodsRank 13 Extending association and prediction models across populations, priors and omics layersGenomicsRank 14 Deep learning infers copy number and tumor-derived content from indirect cancer dataML / AI methodsRank 15 Learned generative models propose biomolecule candidates tested experimentallyStructural bio & protein designRank 16 Comparative genomes tie divergence to adaptation, regulation, and hybrid incompatibilityGenomicsRank 17 Deep models predict mutation, docking, and off-target effects in molecular systemsML / AI methodsRank 18 ML models predict protein binding sites, interface contacts and disordered region functionStructural bio & protein designRank 19 Computational tools target ambient RNA, barcodes, and bulk deconvolution in scRNA-seqSingle-cellRank 20 Dynamical models infer cell-state trajectories and kinetics from single-cell dataSingle-cell
Show 20 more
Rank 21 New pipelines improve strain, metagenome, and pseudogene inference from sequence readsGenomicsRank 22 Learned scoring of structural models to rank molecular interactions across diverse targetsStructural bio & protein designRank 23 Knowledge graphs structure biological associations for target and mechanism discoveryML / AI methodsRank 24 Protocol and evaluation choices shape single-cell expression and regulatory inferenceSingle-cellRank 25 Simulations expose conformational states in disordered, transporter and DNA systemsStructural bio & protein designRank 26 Statistically principled tree estimation and sampling reliability in phylogeneticsGenomicsRank 27 Analytic models show single mechanisms set persistence, extinction and failure boundariesSystems biologyRank 28 Learned models recover cell-state dynamics and count structure from single-cell dataSingle-cellRank 29 LLM agents constrained to versioned tools for reproducible single-cell analysisTools & infrastructureRank 30 Software tools for genotype-to-phenotype prediction, PRS, and variant interpretationTools & infrastructureRank 31 Computational inference of microbial function, interactions, and growth dynamicsSystems biologyRank 32 GPU and algorithmic restructuring speed single-cell, microbiome and metagenomic analysisTools & infrastructureRank 33 Spectra and peptide-level confidence now drive microbial identity and comparisonProteomicsRank 34 Learned generative models sample protein conformations, trajectories and sequencesStructural bio & protein designRank 35 Genomic and ancient protein data resolve hominid, mustelid, and South Asian historiesGenomicsRank 36 Software tools for virome, microbiome, and contaminant detection in sequencing dataTools & infrastructureRank 37 Density-conditioned deep learning recovers particle and conformational heterogeneityStructural bio & protein designRank 38 Latent models isolate behavior-linked and stimulus-specific structure in neural dataSystems biologyRank 39 Mathematical and causal models dissect tumor-immune dynamics in breast cancer therapySystems biologyRank 40 Re-annotation and de novo discovery reveal hidden transposable element diversityGenomics