Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Single-cell
Pathway and program representations are coupled to predictive gene-level inference
Recent preprints pair predictive accuracy with explicit pathway, graph, or gene-program structure for interpretable gene inference.
14 preprints · 1 published · ↑ 2.9× vs 4-wk avg
Rank 2 Tools & infrastructure
Reproducible tools for repeat genotyping, editing readouts, and federated variant sharing
Multi-platform long-read sequencing and federated rare-disease data sharing now demand reproducible, interoperable software beyond single analysis steps.
3 preprints · 2 published · ↑ 2.7× vs 4-wk avg
Rank 3 Single-cell
Spatial atlases and quantitative metrics resolve tissue architecture and rare cells
Spatial omics now pairs reference-scale atlases with metrics that correct size bias and score rare cells, extending tissue-level inference.
5 preprints · 2 published · ↑ 2.2× vs 4-wk avg
Rank 4 ML / AI methods
Clock-like readouts of brain, tissue, and aging state are being built for intervention…
Clock-style biomarkers from EEG, transcriptomes, and fMRI-based brain models now quantify biological state and predict neuromodulation response.
5 preprints · ↑ 2.1× vs 4-wk avg
Rank 5 ML / AI methods
Smartphone video and images yield person-specific mechanical and physiological phenotypes
Low-cost smartphone capture is being paired with learned and physics-constrained models to produce individual-level biomechanical and health outputs.
3 preprints · ↑ 2.0× vs 4-wk avg
Rank 6 Domain priors paired with flow and diffusion generators for DNA and single-cell data↑ 1.8×ML / AI methodsRank 7 Models expose emergent coexistence and coevolutionary structure in ecological systems↑ 1.6×Systems biologyRank 8 Chromosome-level assemblies resolve polyploid origins and karyotype evolution in plants↑ 1.7×GenomicsRank 9 Statistical models refine causal variant, regulatory, and allelic effect inference↑ 1.6×GenomicsRank 10 Long-read methods resolve repeats, mobile elements, methylation and phasing↑ 1.6×GenomicsRank 11 Measuring evidential weight of variant predictors and MAVE assays in classification↑ 1.6×GenomicsRank 12 Packaged tools post-process pathway, m6A, radiation transcriptomic, and cell death data↑ 1.5×Tools & infrastructureRank 13 Computational tools extract single-cell and spatial features from imaging and omics dataSingle-cellRank 14 Benchmarking and patient-level models refine scRNA-seq imputation and stratificationSingle-cellRank 15 Open Python toolkits for isoform, spatial, and single-molecule imaging analysisTools & infrastructureRank 16 Physics-based and deep learning models of cell mechanics and organizationML / AI methodsRank 17 AlphaFold-based predictions extend proteome-scale complex and domain structure coverageStructural bio & protein designRank 18 Compressed dynamic indexes for k-mers, syncmers, and pangenome graphsGenomicsRank 19 Controlled perturbations expose where phylogenetic inference becomes unreliableStructural bio & protein designRank 20 Reduced-order models link mutations and conformational changes to protein dynamicsStructural bio & protein design
Show 20 more
Rank 21 Learned representations of proteins and ligands drive binding and design predictionStructural bio & protein designRank 22 Topological frameworks model higher-order brain network dynamics and interventionsSystems biologyRank 23 Long-read and learned models expand genome-resolved microbial diversity surveysGenomicsRank 24 Reads directly assigned to lineage, genotype, and methylation state without reference…Tools & infrastructureRank 25 Protein language models predict disorder, binding, localization and mutation effectsML / AI methodsRank 26 Deep learning models predict gene, drug and variant effects from biological priorsML / AI methodsRank 27 Protein language model embeddings guide enzyme discovery and structural proximity testsStructural bio & protein designRank 28 Interpretable multi-omics methods link transcriptomic and metabolomic data to functionSystems biologyRank 29 Inferring connectivity distributions from neural dynamics via flow-matching CNFsSystems biologyRank 30 Genome assemblies and metabolic models reveal fungal and bacterial adaptation traitsML / AI methodsRank 31 Physical priors constrain learned models of membrane, peptide, and ligand bindingStructural bio & protein designRank 32 Genome-scale strain diversity linked to disease outcomes and bacteriocin identificationTools & infrastructureRank 33 In silico design of protein binders validated by structure prediction and dockingStructural bio & protein designRank 34 Learned models predict genetic, proteomic, and imaging-derived disease biomarkersML / AI methodsRank 35 Autonomous agents, web platforms and curated databases now support antibody designTools & infrastructureRank 36 Constraint-based genome-scale models paired with learning surrogates and optimisationSystems biologyRank 37 Open-source software tools for single-cell, genome browser, and synteny analysesTools & infrastructureRank 38 Portable, configurable pipelines and GPU basecalling for large-scale sequencing dataTools & infrastructureRank 39 Probabilistic models now infer haplotypes, local ancestry, and cell and tissue lineagesGenomicsRank 40 Benchmarking, spectral library and open-search tools for MS-based molecular analysisProteomics