Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Transcriptomics
Computational methods are resolving spatial cell types, domains, and gene expression
Recent preprints and one publication extend spatial transcriptomics analysis to super-resolution, domain detection, deconvolution, and gene expression modeling.
9 preprints · 1 published · ↑ 3.3× vs 4-wk avg
Rank 2 Genomics
Integrating molecular QTL and GWAS evidence to prioritize disease genes and variants
Recent xQTL, multi-ancestry GWAS and variant effect studies extend noncoding-variant interpretation toward gene prioritization and clinical classification.
4 preprints · 5 published · ↑ 2.2× vs 4-wk avg
Rank 3 ML / AI methods
Biology-aware inference from tumor genomes, cfDNA, and single-cell data
Noninvasive and single-cell tumor measurements are now paired with inference that encodes haplotype, sequence, and cell-state structure.
2 preprints · 5 published · ↑ 2.1× vs 4-wk avg
Rank 4 Genomics
Scalable SNP-based genotype inference for pooled and population genomics data
Pooled single-cell and biobank-scale designs make runtime, memory, and accuracy central constraints for genotype inference.
3 preprints · 3 published · ↑ 1.8× vs 4-wk avg
Rank 5 Genomics
Explicit modeling of sequencing error and read-level evidence in genomic inference
Recent preprints embed sequencing error estimation and nanopore dwelling-time signals directly within calling and calibration workflows.
5 preprints · 1 published · ↑ 1.8× vs 4-wk avg
Rank 6 Pretrained embeddings localize binding residues and transfer across peptide chirality↑ 1.7×Structural bio & protein designRank 7 Statistical controls for artifacts, batch effects and isoforms in expression analysis↑ 1.7×TranscriptomicsRank 8 Generative models couple cryo-EM and cryo-ET data to biomolecular structure↑ 1.7×Structural bio & protein designRank 9 Open tools unify heterogeneous biochemical and omics data into queryable annotated form↑ 1.5×Tools & infrastructureRank 10 Computational mapping of noncoding variants to target genes and regulatory elementsEpigenomicsRank 11 Learned and explicitly modelled components reshape mass-spec proteomics analysisProteomicsRank 12 Learned models forecast single-cell gene expression and perturbation responsesSingle-cellRank 13 Gene content and genome organization shape lineage-specific trait evolution across taxaGenomicsRank 14 Sequence-based models and pipelines infer pathogen resistance and infection phenotypesTools & infrastructureRank 15 Learned representations predict drug combinations, interactions, and molecular propertiesML / AI methodsRank 16 TE activity shapes LINE-1 chimeras, fungal transfer, and maize centromere repeatsGenomicsRank 17 Sequence and motif geometry methods probe protein and RNA structure inferenceStructural bio & protein designRank 18 Reproducible pipelines turn microbial sequence data into interpretable biological callsTools & infrastructureRank 19 Language models learn DNA, RNA and protein sequence representations for predictionML / AI methodsRank 20 Single-cell atlases paired with machine learning resolve rare states and lineage structureSingle-cell
Show 20 moreShow less
Rank 21 Evaluation design, not model class, limits predictive claims in biological MLML / AI methodsRank 22 Local cell interactions drive collective invasion and tissue-scale polarity in modelsSystems biologyRank 23 Computational screens and designs prioritizing candidates for experimental testingStructural bio & protein designRank 24 Knowledge graphs and benchmarks evaluate computational drug and resistance predictionML / AI methodsRank 25 Neural and hybrid models infer population history from LD, ARGs, and mtDNA distancesGenomicsRank 26 Configuration-driven tools build local relational databases for omics outputsTools & infrastructureRank 27 MD simulations resolve LNP-siRNA structure, PNPLA membrane binding, and PET hydrolysisStructural bio & protein designRank 28 Software converts raw time-resolved microscopy into quantified cellular dynamicsTools & infrastructureRank 29 Learned models infer post-mortem interval and circadian time from molecular profilesMetabolomicsRank 30 Decomposition and restricted computation cut runtime of core sequence inferenceTools & infrastructureRank 31 Computational models resolve senescent and low-complexity cell states from scRNA dataSingle-cellRank 32 Learned models integrate spatial transcriptomics, pathology, and single-cell omics…Single-cellRank 33 Mechanistic models link regulatory dynamics to cell states and signaling phenotypesSystems biologyRank 34 Web tools predict siRNA toxicity and catalog peptide-mediated siRNA delivery dataTools & infrastructureRank 35 Cancer genome studies classify mutation function, signatures and oncogene amplificationGenomicsRank 36 Latent-variable and local-learning models account for visual cortical responsesSystems biologyRank 37 Metabolic and energy budgets shape neural coding and synaptic precisionML / AI methodsRank 38 PCA-compressed Euclidean alignment drives fast 2D classification of cryo-EM particlesStructural bio & protein designRank 39 Learned models predict conformational ensembles, antigenic change and nanobody foldingStructural bio & protein designRank 40 ML tools for isomiR-aware target prediction and artificial miRNA designML / AI methods