Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Single-cell
Computational methods infer spatial domains and cell types from spatial omics data
Multiplexed imaging and Xenium-scale spatial profiling demand interpretable methods for domain and cell-type inference.
9 preprints · 3 published · ↑ 3.3× vs 4-wk avg
Rank 2 Genomics
Reference-free and hybrid-data methods infer genetic variation from sequencing data
Recent preprints and publications extend variant calling, phasing, and population-size inference to sparse, pooled, and reference-free sequencing data.
7 preprints · 5 published · ↑ 3.2× vs 4-wk avg
Rank 3 Genomics
Integrating xQTL and GWAS evidence to prioritize genes and interpret variants
Multi-ancestry GWAS and xQTL integration now link noncoding variants to druggable and disease genes across several diseases.
3 preprints · 5 published · ↑ 2.9× vs 4-wk avg
Rank 4 Structural bio & protein design
Learned representations of protein pockets and peptides predict binding and design…
Recent preprints and papers apply learned protein and peptide encoders to binding prediction, pocket mapping, and generative design across diverse targets.
5 preprints · 4 published · ↑ 2.7× vs 4-wk avg
Rank 5 Structural bio & protein design
Learned models infer biomolecular structure and antigenic change from embeddings
Recent preprints and papers apply pretrained language and generative models to prediction and design across proteins, antibodies and viruses.
7 preprints · 3 published · ↑ 2.4× vs 4-wk avg
Rank 6 Structure-aware inference links cancer mutations, chromatin marks and cell states↑ 2.2×EpigenomicsRank 7 Biology-informed representations for predicting microbial phenotypes and enzyme function↑ 2.3×ML / AI methodsRank 8 Validation design now decides whether single-cell prediction models hold up↑ 2.1×Single-cellRank 9 Benchmarks Test Whether Pretrained Biological Embeddings Improve Downstream Prediction↑ 1.8×ML / AI methodsRank 10 Resistance Determinants Detected Directly From Raw Sequencing Reads Without Assembly↑ 1.7×Tools & infrastructureRank 11 Single-cell profiling links human aging, lineage history and immune variation↑ 1.8×Single-cellRank 12 ML models prioritize GPCR and TDP-43 candidates before experimental testing↑ 1.6×ML / AI methodsRank 13 LLM agents and knowledge graphs build provenance-traced biomedical evidence networks↑ 1.7×Single-cellRank 14 Inferring pathogen identity, resistance, and variant composition from indirect genomic…↑ 1.5×GenomicsRank 15 Computational models infer mechanism from indirect, heterogeneous biological data↑ 1.6×Systems biologyRank 16 Explicit error modelling and learned components drive modern omics identificationMetabolomicsRank 17 Physics-based and generative methods advance protein-ligand binding and molecular designStructural bio & protein designRank 18 Computational methods model RNA folding, 3D motifs, and PIWI piRNA cleavage specificityStructural bio & protein designRank 19 Error control and artifact removal reshape transcriptome expression analysisTranscriptomicsRank 20 Cryo-EM and cryo-ET density data constrain generative models of atomic structureStructural bio & protein design
Show 20 more
Rank 21 Local, configuration-driven frameworks structure omics data for reproducible reuseTools & infrastructureRank 22 Faster optimal and compressed alignment methods for graph and pangenome sequence dataTools & infrastructureRank 23 Computational tracing of transposable element activity and repeat architectureGenomicsRank 24 Comparative and community-curated annotation of eukaryotic genomesGenomicsRank 25 Fast RNA folding engines coupled to codon design and single-sequence structure predictionML / AI methodsRank 26 Fused multimodal representations predict drug and protein interaction relationshipsML / AI methodsRank 27 Read-level inference and genome atlases resolve host and function in microbiomesSystems biologyRank 28 Models link microbial growth, resource limits, and antimicrobial treatment outcomesSystems biologyRank 29 New open tools for alignment analysis, regulatory modelling, knockout phenotype networksTools & infrastructureRank 30 Comparative genomics links gene loss and synteny to lineage-specific trait evolutionGenomicsRank 31 Cancer genome datasets resolve mechanisms of amplification, two-hit loss and signaturesGenomicsRank 32 ML-guided design of artificial microRNAs and functional consequences of miRNA gene…GenomicsRank 33 Directed connectivity and criticality scaling in resting-state brain activitySystems biologyRank 34 Simulations resolve membrane-dependent lipid pKa, PDIM sorting and polymyxin penetrationStructural bio & protein designRank 35 Quantitative models infer circadian time, affective switching and PKA dynamicsSystems biologyRank 36 MD simulations map energetic couplings in channel gating and peptide membrane entryStructural bio & protein designRank 37 Learned models predict ADMET, peptide retention, P-gp transport, and pancreatic cancerML / AI methodsRank 38 Computational models nominate aging, metabolic, and Long COVID targets from network dataSystems biologyRank 39 Packaged workflows impose explicit statistical structure on high-throughput omics dataML / AI methodsRank 40 Machine learning respecting pathology and expert labels in clinical brain dataML / AI methods