Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Tools & infrastructure
Knowledgebases link variants and regulatory elements to trait genetics and cell states
Expanded resources such as the ENCODE cCRE registry (2.37M human elements) and CIMA (10M+ immune cells) now span regulatory and trait-linked genetics.
1 preprint · 22 published · ↑ 5.2× vs 4-wk avg
Rank 2 Genomics
Curated databases standardize microbial genomes, metagenomes and metadata across habitats
Rapid growth of metagenomic sequencing, as gcMeta 2025 notes, makes harmonized annotation and metadata essential for reuse across studies.
20 published · ↑ 4.2× vs 4-wk avg
Rank 3 Transcriptomics
Methods infer spatial tissue organization and cell identity from spatial transcriptomics
Recent preprints and publications extend histology-guided deconvolution, 3D reconstruction, and spatial domain identification for spatial transcriptomics.
5 preprints · 23 published · ↑ 3.6× vs 4-wk avg
Rank 4 Tools & infrastructure
Expanded databases map noncoding RNA regulation across bulk and single-cell datasets
Updated resources now pair noncoding RNA and regulatory annotations with single-cell, spatial, and pan-cancer profiles at larger scale.
2 preprints · 22 published · ↑ 3.4× vs 4-wk avg
Rank 5 ML / AI methods
Fusing graph, language-model and structure embeddings for drug-target prediction
Multimodal architectures now jointly model several molecular representations across drug-target, drug-drug, drug-gene, peptide and binding-site tasks.
4 preprints · 11 published · ↑ 3.0× vs 4-wk avg
Rank 6 Curated databases link bioactivity records to structural and mechanistic annotation↑ 3.0×Tools & infrastructureRank 7 Curated databases re-integrate public omics data for stress, drug, and cell-cell biology↑ 2.1×Tools & infrastructureRank 8 Computational models tie formulations, peptides and media to experimental readouts↑ 2.3×ML / AI methodsRank 9 Statistical methods recover information lost to dropout and unresolved cell composition↑ 2.3×Single-cellRank 10 Web atlases integrate single-cell and spatial transcriptomes across species and tissues↑ 1.9×Tools & infrastructureRank 11 Graph models fuse network priors with omics to infer regulation and driver genes↑ 2.0×ML / AI methodsRank 12 Structure-aware graph encodings guide binding, GPCR activity and molecular generation↑ 2.0×Structural bio & protein designRank 13 Curated databases unify viral genomes, metadata, and immune receptor reference sequences↑ 2.0×Tools & infrastructureRank 14 Reference and population context increasingly shape variant detection and inference↑ 1.8×GenomicsRank 15 Scaffolding and gap filling target residual fragmentation in long-read assemblies↑ 2.0×GenomicsRank 16 Curated metabolic databases add programmatic access and expanded reaction coverage↑ 1.8×Tools & infrastructureRank 17 Curated databases consolidate circRNA, RNA modification and polyA site data across species↑ 1.8×Tools & infrastructureRank 18 Deep learning links routine cancer histology and genomics to molecular biomarkers↑ 1.7×ML / AI methodsRank 19 Inferred 3D chromosome structures resolve single-cell and time-resolved chromatin states↑ 1.8×EpigenomicsRank 20 Learned and simulated models probe protein conformational dynamics and ligand binding↑ 1.5×Structural bio & protein design
Show 20 more
Rank 21 Structure-informed representations are sharpening remote homology and annotation searches↑ 1.5×Structural bio & protein designRank 22 Single-cell models are being checked by benchmarks and information-theoretic boundsSingle-cellRank 23 Machine learning models predict TCR, HLA and host-pathogen protein recognitionML / AI methodsRank 24 Learned graph and sequence encodings predict drug, peptide and antimicrobial activityML / AI methodsRank 25 Ligand-resolved kinetic models map small-molecule control of conformational equilibriaStructural bio & protein designRank 26 Predictive kinetic and thermodynamic models replace empirical tuning in molecular designGenomicsRank 27 Neural and Bayesian models trained on simulated data infer biological parametersML / AI methodsRank 28 Synaptic wiring alone assigns neuronal types; microscopy segmentation tools advanceML / AI methodsRank 29 Sequence and chromatin context both constrain transcription factor bindingEpigenomicsRank 30 Multi-ancestry and rare-variant frameworks refine genetic association and fine-mappingGenomicsRank 31 Ligand-aware structure prediction and generative screening target binding-mode accuracyStructural bio & protein designRank 32 Causal and mediation methods extend omics analysis beyond linear, homogeneous settingsSystems biologyRank 33 Proteomics resources expand reference proteomes, abundance data, and FAIR data sharingProteomicsRank 34 MD simulations resolve conformational landscapes governing inhibitor binding and gatingStructural bio & protein designRank 35 Stochastic variability and network structure govern timing of cellular responsesSystems biologyRank 36 Learned predictors of conformational ensembles are bounded by training distributionsStructural bio & protein designRank 37 Proteome-scale analyses link sequence statistics and structural dynamics to constraintGenomicsRank 38 Learned spectrum models for peptide and metabolite identification in tandem MSProteomicsRank 39 Gap-free assemblies and pangenomes expose structural variation across crop lineagesGenomicsRank 40 Computational models infer chromatin structure, molecular motion and cell-state dynamicsEpigenomics