Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Single-cell
Validated, reproducible pipelines for single-cell and spatial omics analysis
Recent preprints add explicit validation of normalization, correction, and clustering steps to single-cell and spatial omics workflows.
13 preprints · 1 published · ↑ 2.6× vs 4-wk avg
Rank 2 ML / AI methods
Protein language models adapted to targets with sequence-only and interpretable designs
Recent preprints test whether protein language model embeddings can be made interpretable and task-specific without requiring structure input.
8 preprints · 1 published · ↑ 2.0× vs 4-wk avg
Rank 3 ML / AI methods
Biological priors and interpretability probes in omics and pathology deep learning
Recent preprints and published methods pair prior pathway and network knowledge with interpretable outputs across omics, cytometry, and histopathology.
7 preprints · 3 published · ↑ 1.9× vs 4-wk avg
Rank 4 Structural bio & protein design
Learned correction and scoring of predicted biomolecular complexes and interactions
Structure predictors now generate many candidate complexes, so correction, scoring, and selection models are needed to rank them.
5 preprints · 4 published · ↑ 1.9× vs 4-wk avg
Rank 5 Single-cell
Benchmarks probe whether genomic and single-cell models generalize beyond training
Learned genomic and single-cell models are now judged on mechanism, intervention geometry and robustness rather than raw predictive accuracy.
9 preprints · ↑ 1.8× vs 4-wk avg
Rank 6 Cell division history explains more epigenetic noise than chronological age↑ 1.7×GenomicsRank 7 Structural priors link latent embeddings to spatial tissue and chromatin organization↑ 1.5×ML / AI methodsRank 8 Inferring transcriptional state and regulatory networks from single-cell and bulk dataSingle-cellRank 9 Learned models evaluated against fixed baselines across prognosis and drug discoveryML / AI methodsRank 10 Genetic association methods address confounding, pleiotropy, and feature interactionsGenomicsRank 11 ML models address potency, ADMET and toxicity decision points in molecular designML / AI methodsRank 12 Computational tools recover splicing and isoform structure from RNA-seq and long readsGenomicsRank 13 Structure predictors fail on localized protein subsets, not just in aggregate accuracyStructural bio & protein designRank 14 Model component choice shapes simulated neural outcomes in EEG, DBS, and cortex modelsSystems biologyRank 15 Compact CpG selection and memory-limited active learning for predictive biochemistryEpigenomicsRank 16 Deep learning predictors for genes, phage hosts and RNA face generalization testsGenomicsRank 17 Chromosome-scale assemblies resolve haplotypes and subgenomes in plants and a fungusGenomicsRank 18 Evaluation design now governs reported performance of ML molecular property modelsML / AI methodsRank 19 Computational inference of cell identity and fate across tissues from single-cell dataSingle-cellRank 20 Inference for stochastic and dynamical mechanistic models in systems biologySystems biology
Show 17 more
Rank 21 Refined force fields and generative models sample disordered and nucleic acid statesStructural bio & protein designRank 22 Learned models predict RNA yield, 3'UTR stability, and bacterial sRNA-mRNA outcomesML / AI methodsRank 23 MD ensembles show glycosylation and membrane curvature reshape conformational landscapesStructural bio & protein designRank 24 Reference content, measurement design, and target choice shape proteomic readoutsProteomicsRank 25 Frozen and foundation encoders lose robustness under acquisition and corruption shiftsML / AI methodsRank 26 Standalone tools reanalyze AlphaFold 3 and ColabFold outputs for interface scoringTools & infrastructureRank 27 Micro-scale rules recovered as tissue-level descriptions in cell population modelsSystems biologyRank 28 Scaling phylogenetic network and species-tree inference beyond bifurcating treesGenomicsRank 29 Sequence models predict accessibility and design tissue-specific enhancers in mouseEpigenomicsRank 30 Reference representation shapes variant detection in pangenome and population settingsGenomicsRank 31 Quantified imputation of missing values in single-cell RNA-seq under dropoutSingle-cellRank 32 Chromosome-level and marker-based resources reveal how assembly design shapes inferenceGenomicsRank 33 AI-designed peptides and engineered nucleases assessed by biological readoutsStructural bio & protein designRank 34 Nonlinear feedback yields hysteresis and multistability in network spreading modelsSystems biologyRank 35 Anomaly detection and reconstruction-aware pipelines reshape cryo-EM particle pickingStructural bio & protein designRank 36 Generative models design compounds, PROTAC linkers, and retrosynthetic SMILES fragmentsStructural bio & protein designRank 37 Enhanced sampling and force-field corrections refine RNA and POLG conformational modelsStructural bio & protein design