Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Single-cell
Models infer regulatory networks from single-cell and spatial transcriptomic data
Integrating scRNA-seq with spatial and chromatin data now enables inference of regulatory and signaling networks across tissue contexts.
14 preprints · 5 published · ↑ 4.0× vs 4-wk avg
Rank 2 Genomics
Mutational, regulatory and structural features link genomic instability to cancer
Single-cell, CRISPR and regulatory profiling now resolve how instability, repair and regulatory variation shape tumor genomes.
3 preprints · 3 published · ↑ 3.0× vs 4-wk avg
Rank 3 Genomics
Null-model bias and molecular QTL mapping sharpen causal inference in complex traits
Large cohorts and multi-omic atlases now enable genetic mapping of expression noise, isoforms, and metabolites alongside biases in GWAS null statistics.
2 preprints · 6 published · ↑ 2.9× vs 4-wk avg
Rank 4 ML / AI methods
Aligning unpaired omics modalities with optimal transport and contrastive learning
Recent preprints apply optimal transport and graph geometry to align spatial and single-cell modalities that lack paired measurements or shared features.
11 preprints · 3 published · ↑ 2.8× vs 4-wk avg
Rank 5 ML / AI methods
Deep learning models predict molecular interactions from learned representations
Recent preprints and publications pair graph, transformer, and image-based encoders with task-specific heads for interaction and binding prediction.
3 preprints · 4 published · ↑ 2.2× vs 4-wk avg
Rank 6 Network topology and sequence graphs are becoming objects of inference in genomics↑ 1.9×GenomicsRank 7 Structural and diversity metrics reveal animal genome evolution across lineages↑ 1.9×GenomicsRank 8 Preprocessing and benchmarking workflows for spatial, single-cell, and proteomic data↑ 1.7×Tools & infrastructureRank 9 Penalized and goal-directed models stratify omics data for vaccine and sepsis outcomes↑ 1.8×Tools & infrastructureRank 10 New computational methods refine gene regulatory and transcript-level analyses↑ 1.6×Systems biologyRank 11 Benchmarking sequencing and informatics tools against reference data across platforms↑ 1.5×GenomicsRank 12 Isoform-resolved long-read data refine splicing, NMD rules and haplotype proteomes↑ 1.5×GenomicsRank 13 Language models now drive protein, mRNA, and codon sequence design and predictionML / AI methodsRank 14 Calibrating and detecting variant effects and copy numbers from sequencing dataGenomicsRank 15 Probabilistic and deep learning models infer missing or latent omics measurements↑ 1.6×EpigenomicsRank 16 Tools refine genome assembly assessment, metagenomic abundance, and amplicon taxonomyTools & infrastructureRank 17 Models link somatic mutation, plasticity, and spatial aging to disease dynamics↑ 1.6×Systems biologyRank 18 Experimental restraints and coarse-grained models resolve protein conformational ensemblesStructural bio & protein designRank 19 Sampling frames define inference scope in wastewater, serology and vaccine trial dataML / AI methodsRank 20 Language models and optimization steer RNA sequence design toward structure and binding↑ 1.5×Structural bio & protein design
Show 20 moreShow less
Rank 21 Compressed indexes and sketches speed k-mer and sequence similarity search at scaleTools & infrastructureRank 22 Structure-based computational design of peptide, enzyme, and antibody variantsStructural bio & protein designRank 23 Brain connectivity beyond pairwise links: higher-order and manifold representationsSystems biologyRank 24 Pretrained embeddings fused with structural or graph features for binding predictionML / AI methodsRank 25 Stochastic cell models link molecular noise and division control to population behaviorSystems biologyRank 26 Motif-level models interpret cis-regulatory syntax and predict miRNA-mRNA interactionsML / AI methodsRank 27 Phase-plane and agent-based models frame disease onset as cell-level state transitionsSystems biologyRank 28 Topological and network-embedding methods map Alzheimer's disease molecular structureSystems biologyRank 29 Diffeomorphic and velocity-field registration map whole-brain data onto mouse atlasesStructural bio & protein designRank 30 Automated pipelines chain reproducible steps toward drug and biomarker endpointsTools & infrastructureRank 31 Learned models address missing and heterogeneous omics data across platforms and tissuesML / AI methodsRank 32 Thermodynamic and kinetic constraints tightening metabolic models against measured dataSystems biologyRank 33 ML and enhanced sampling accelerate structure-based fragment and virtual screeningStructural bio & protein designRank 34 Pattern-based tools map DNA methylation motifs and Z-DNA forming sequences genome-wideEpigenomicsRank 35 Formal algorithms for non-tree genomic structures in pangenome and phylogenetic networksGenomicsRank 36 Sequence-based inference links smORFs, disordered regions and novel enzymes to functionStructural bio & protein designRank 37 Convergent antibody binding modes and near-universal SARS-CoV-2 immune escapeStructural bio & protein designRank 38 Lipids, ions and electrostatics tune transporter conformational energeticsStructural bio & protein designRank 39 LLM agents turn literature, structures and notebooks into biological hypothesesML / AI methodsRank 40 Computational separation of protein information from pooled or confounding proteomic dataProteomics