Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Single-cell
Generative and graph models infer cell states and spatial gene structure
Generative and graph-based models now address alignment, annotation, imputation and reconstruction in single-cell and spatial data.
14 preprints · 6 published · ↑ 3.8× vs 4-wk avg
Rank 2 ML / AI methods
Pretrained embeddings with deep networks advance molecular interaction prediction
Recent work fuses ESM and ChemBERTa embeddings with 3D and graph features, adding contrastive pretraining and modification-aware benchmarks for binding and…
9 preprints · 5 published · ↑ 3.4× vs 4-wk avg
Rank 3 Transcriptomics
Modeling technical variability before biological inference in sequencing data
Recent preprints and publications explicitly model normalization, scale, and quantification uncertainty in ChIP-seq, RNA-seq, and transcript-level analyses.
3 preprints · 2 published · ↑ 2.7× vs 4-wk avg
Rank 4 ML / AI methods
LLMs and deep models for variant, protein, and trial outcome prediction under scrutiny
Recent preprints test whether genomic and clinical ML predictors are calibrated, interpretable, and robust to confounders and leakage.
9 preprints · 2 published · ↑ 2.4× vs 4-wk avg
Rank 5 Genomics
Graph-aware and callable-site methods reformulate standard population-genetic inference
Population-scale sequence data increasingly demand graph-aware and callable-site-aware tools for genotyping, selection, and introgression.
4 preprints · 3 published · ↑ 2.2× vs 4-wk avg
Rank 6 Conditional language models steer biological sequence design toward defined targets↑ 2.2×ML / AI methodsRank 7 Genome-wide readouts are linking cancer origin, DNA repair and genomic instability↑ 2.2×Systems biologyRank 8 Stochastic molecular rules set population-level expression distributions and noise↑ 2.3×Systems biologyRank 9 Input representation reshapes DNA language models for variant and sequence tasks↑ 2.0×ML / AI methodsRank 10 Lipids and ligands reshape conformational energy landscapes in membrane proteins↑ 1.8×Structural bio & protein designRank 11 Parallel and out-of-core software accelerates genomic read and interval processing↑ 1.9×Tools & infrastructureRank 12 Spatial and single-cell profiling maps prognostic tumor niches and cell states↑ 1.6×TranscriptomicsRank 13 Deep learning maps tissue morphology to spatial omics and clinical endpoints↑ 1.7×ML / AI methodsRank 14 Structure-based computation guides ligand, antibody, and GPCR design decisions↑ 1.7×Structural bio & protein designRank 15 Simulation-based inference tackles intractable likelihoods in epidemic and phylodynamic…↑ 1.8×Systems biologyRank 16 Latent-factor models recover interpretable structure from multi-omics and compositional…↑ 1.6×Systems biologyRank 17 Fitted models are validated against independent neural measurements across scales↑ 1.6×ML / AI methodsRank 18 Identifiability and parameter estimation methods for mechanistic biological models↑ 1.6×Systems biologyRank 19 Ensemble models and force fields are tied to SAXS, NMR, and ITC observablesStructural bio & protein designRank 20 Coupling mechanical loading to multiscale models of tissue and matrix behaviourSystems biology
Show 20 more
Rank 21 Calibrated benchmarks guide method choice in protein structure and library analysisTools & infrastructureRank 22 Robust models resolve protein abundance from noisy mass spectrometry measurementsProteomicsRank 23 k-mer and embedding methods refine taxonomic classification of genomic sequencesGenomicsRank 24 Harmonized antibody-antigen data meets interface-level structural analysis toolsTools & infrastructureRank 25 Methods refine calibrated, context-resolved genetic association and fine-mappingGenomicsRank 26 Long-read and structural context are resolving isoform-level regulatory annotationGenomicsRank 27 Scalable algorithms for viral discovery, read alignment, and amplicon and genome analysisTools & infrastructureRank 28 Structure and sequence models predict nanopore readouts for proteins and XNA basesGenomicsRank 29 Quantitative models link mutation fitness effects to genome and virulence evolutionGenomicsRank 30 Pangenome graphs and bias correction methods validated on large sequencing datasetsGenomicsRank 31 Computational tools for isotope flux, microbiome, and microbial interaction modelingMetabolomicsRank 32 Graph models link expression data to pathway activity and perturbation responsesSingle-cellRank 33 Knowledge-guided and contrastive models integrate heterogeneous single-cell dataSingle-cellRank 34 Protein language model embeddings reveal motifs, local structure, and remote homologyML / AI methodsRank 35 Assessing assumptions and uncertainty in phylogenetic and phylodynamic inferenceSystems biologyRank 36 Benchmarks and adaptations of computational methods for RNA and cell-free DNA omics dataTranscriptomicsRank 37 Computational methods for interpreting genomic variants from sequencing dataGenomicsRank 38 Physics-informed models reconstruct cell population dynamics from single-cell dataSingle-cellRank 39 Embedding and retrieval conditioning move from protein search into generative designStructural bio & protein designRank 40 Negative-set design and data construction govern binding prediction reliabilityML / AI methods