Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Single-cell
Single-cell and sequencing platforms validated against matched or reference data
Recent preprints and publications benchmark sequencing platforms and single-cell methods against matched FFPE tissue, bulk data, or reference annotations.
10 preprints · 3 published · ↑ 3.4× vs 4-wk avg
Rank 2 ML / AI methods
Protein language models and graph networks predict enzyme and protein properties
Recent preprints and papers pair protein language model embeddings with explainable or LLM-based components for enzyme, solubility, and molecular tasks.
5 preprints · 3 published · ↑ 2.9× vs 4-wk avg
Rank 3 Single-cell
Latent models reconstruct cell fate trajectories and predict perturbation responses
Foundation models and transport methods for cell dynamics are now benchmarked against simple baselines and tested on unseen contexts.
14 preprints · 4 published · ↑ 2.7× vs 4-wk avg
Rank 4 Epigenomics
Single-cell and multiome epigenomic readouts feed new regulatory mapping models
Single-cell and multiome chromatin profiles now feed deep-learning and statistical models that resolve context-specific regulatory elements and gene expression.
5 preprints · 4 published · ↑ 2.4× vs 4-wk avg
Rank 5 Genomics
Learned models increasingly substitute for homology in genome and protein annotation
Protein language model embeddings and genomic language models now address annotation tasks where homology is absent or unreliable.
5 preprints · 1 published · ↑ 2.4× vs 4-wk avg
Rank 6 Local ancestry dosages and ancestry-aware models improve genotype-based inference↑ 2.2×GenomicsRank 7 Physics-based binding free energies and learned models for fragment and peptide ligands↑ 2.0×Structural bio & protein designRank 8 Bacterial genomic context resolves plasmids, CRISPR interactions and transmission↑ 1.9×GenomicsRank 9 Learned models fuse single-cell, bulk, spatial, and imaging data for prediction↑ 1.9×Single-cellRank 10 Mosaic mitochondrial and tandem repeat variation quantified from sequencing data↑ 1.9×GenomicsRank 11 LLM agents orchestrate existing protein design tools into executable workflows↑ 1.6×Tools & infrastructureRank 12 Generative models learn conformational ensembles of proteins, peptides, and complexes↑ 1.5×Structural bio & protein designRank 13 Protein and RNA language model embeddings fused with auxiliary data for predictionML / AI methodsRank 14 Deep learning transfers drug-response models from cell lines toward patient tumorsML / AI methodsRank 15 Neighbourhood graphs underpin outlier trimming and clustering in single-cell omicsSingle-cellRank 16 Network and graph models are ranking causal genes from GWAS, QTL and expression dataGenomicsRank 17 Coupling histology and expression latents for spatial tissue and 3D reconstructionTranscriptomicsRank 18 Simulation models link timing, migration and structure to evolutionary outcomesSystems biologyRank 19 Probabilistic models span genomic data integration, copulas and network inferenceML / AI methodsRank 20 Metapopulation models couple mobility and surveillance structure to pathogen dynamicsSystems biology
Show 20 more
Rank 21 Computational models resolve ecDNA cycles, replication timing and guide efficiencyML / AI methodsRank 22 MD and electrostatics test whether predicted protein structures get charge states rightStructural bio & protein designRank 23 Learned sequence and geometric models for peptide-MHC binding and RNA regulatory designML / AI methodsRank 24 Multi-omic tissue profiling resolves cell states, glycoproteomes and methylation patternsEpigenomicsRank 25 Reliability checks for omics ML across artifact filtering, leakage control, and curationTools & infrastructureRank 26 Quality-aware tools correct artefacts and resolve strains in viral sequencing dataTools & infrastructureRank 27 ML pipelines extract quantitative readouts from microscopy and spatial tissue dataML / AI methodsRank 28 LLM agents are automating reproducible single-cell and spatial omics analysis workflowsML / AI methodsRank 29 Structural and tissue-context models refine variant and regulatory interpretationStructural bio & protein designRank 30 Haplotype-resolved inference from long reads, Hi-C and biobank haplotypesGenomicsRank 31 MD ensembles link conformational states to catalysis, release kinetics, and allosteryStructural bio & protein designRank 32 Parallel and hardware-tuned tools speed up genomic and phylogenetic analysesTools & infrastructureRank 33 Quantitative models and tools formalize the dynamics of cellular processesSystems biologyRank 34 Structure prediction and generative models drive binder and framework designStructural bio & protein designRank 35 Computational readouts from fluorescence microscopy for subcellular quantificationSingle-cellRank 36 Open-source software packaging image analysis pipelines for biomedical and plant imagingTools & infrastructureRank 37 Generative models map protein fitness, backbone geometry, and replica exchange samplingStructural bio & protein designRank 38 Personalized computational models of heart, lung, and musculoskeletal systemsSystems biologyRank 39 Computational models of receptor energetics, binding and membrane-scale behaviorStructural bio & protein designRank 40 Sequence data and learned representations assign disease and resistance labelsGenomics