Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Genomics
Isoform-aware models and pipelines annotate transcripts and regulatory networks
Long-read RNA-seq and computational models now resolve transcript isoforms, splicing regulators, and expression noise at population scale.
4 preprints · 6 published · ↑ 3.5× vs 4-wk avg
Rank 2 Epigenomics
Sequence-trained models predict methylation and enhancers while exposing motif syntax
Recent preprints extend sequence-based prediction to methylation, enhancers and cis-regulatory syntax, with interpretable attribution of model drivers.
10 preprints · 1 published · ↑ 2.8× vs 4-wk avg
Rank 3 Single-cell
Unpaired and spatial omics modalities aligned via optimal transport and graph geometry
Foundation models and spatial assays now produce heterogeneous embeddings that must be aligned across unpaired or out-of-distribution modalities.
8 preprints · 6 published · ↑ 2.7× vs 4-wk avg
Rank 4 Tools & infrastructure
Computational integration of microbial omics with structural and metabolic context
Protein language models, structural annotation, and taxon-resolved metrics now resolve microbial function and dysbiosis from omics and sequencing data.
3 preprints · 7 published · ↑ 2.4× vs 4-wk avg
Rank 5 Genomics
Polymer models, reference genomes and DNA shape reveal genome structure and variation
Recent tools and references link 3D chromatin structure, cell line-matched genomes and structural variant mechanisms to specific genomic readouts.
1 preprint · 3 published · ↑ 2.3× vs 4-wk avg
Rank 6 Reference-anchored models for deconvolution, lineage, and velocity in single-cell data↑ 2.2×Single-cellRank 7 Language models now design and predict RNA and mRNA sequence function↑ 2.0×ML / AI methodsRank 8 Omics models stratify disease and predict vaccine reactogenicity from transcriptomes↑ 1.9×Tools & infrastructureRank 9 Fitting plasticity rules and circuit architecture in neural models to behavioral data↑ 1.9×Systems biologyRank 10 Learned models impute and integrate omics data across platforms, tissues, and missingness↑ 1.9×ML / AI methodsRank 11 Epistatic omission biases GWAS nulls as isoform and non-coding variant mapping advances↑ 1.9×GenomicsRank 12 Chromosome fusions, genome-wide coupling and ion channel diversity across animals↑ 1.9×GenomicsRank 13 Simulations show how lipids, peptides, and mutations reshape protein assembly states↑ 1.8×Structural bio & protein designRank 14 Reference-based benchmarks test sequencing and proteomics informatics tools↑ 1.6×Tools & infrastructureRank 15 Graph and latent models of EEG, fMRI and connectome data for brain connectivity↑ 1.9×ML / AI methodsRank 16 Protein language model embeddings predict binding, epitopes, and designed sequences↑ 1.7×Structural bio & protein designRank 17 Local stochastic rules shape collective outcomes in chemotaxis and membrane proteins↑ 1.5×Systems biologyRank 18 Benchmarking and input design shape phylogenetic inference accuracyGenomicsRank 19 Bistable models link cellular state transitions to disease onset and tissue injury↑ 1.6×Systems biologyRank 20 Learned models sharpen variant, copy number, and splice calls from sequencing dataGenomics
Show 20 moreShow less
Rank 21 Population-scale tools for ancestry, haplogroup and linkage disequilibrium inferenceGenomicsRank 22 Methods for gene specificity and co-expression in single-cell and spatial transcriptomicsTranscriptomicsRank 23 Single-cell and bulk genomic data used to infer regulatory variants and networksGenomicsRank 24 Structure-guided generative design of binders, hybrid proteins, and antibody librariesStructural bio & protein designRank 25 Simulations map Wnt receptor trafficking and breast cancer plasticity dynamicsSystems biologyRank 26 Molecular graph networks fused with multimodal inputs for drug response predictionML / AI methodsRank 27 Compressed indexes and banded alignment reduce memory in large sequence analysesTools & infrastructureRank 28 Computational methods for interaction mapping, tissue imaging and EV protein tracingProteomicsRank 29 Interpretability and debiasing scrutinize sequence-based bioactivity predictionML / AI methodsRank 30 Multi-omic data and thermodynamic constraints refine metabolic network modelsSystems biologyRank 31 Nominating candidate targets and biomarkers from multi-source genomic and clinical dataTools & infrastructureRank 32 Pretrained protein embeddings fused with auxiliary features for site predictionML / AI methodsRank 33 Topological and higher-order representations move beyond pairwise biological networksSystems biologyRank 34 Structured models reconciling heterogeneous data across labs, species, and time pointsSystems biologyRank 35 Convergent SARS-CoV-2 antibody paratopes and near-universal variant escapeStructural bio & protein designRank 36 Generative and velocity-field registration map mouse brain data onto common atlasesTools & infrastructureRank 37 Productive reiterative initiation and de novo microproteins in enterobacteriaTranscriptomicsRank 38 MD and cryo-EM reveal lipid and anion control of transporter conformational barriersStructural bio & protein designRank 39 NMR-guided simulations resolve dynamics of disordered and partly folded proteinsStructural bio & protein designRank 40 Reproducible pipelines extend from network disease modules to docking and LLM gene setsTools & infrastructure