Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Single-cell
Computational methods infer tissue organization from spatially resolved transcriptomic…
New spatial methods target domain discovery, marker detection, cell annotation and benchmarking on imaging and sequencing platforms.
10 preprints · 2 published · ↑ 2.7× vs 4-wk avg
Rank 2 Genomics
Reference-free and hybrid-data methods refine variant calling across sequencing platforms
Long-read dwelling-time data, hybrid long-short reads, and pooled-design pipelines are extending variant detection beyond standard short-read workflows.
4 preprints · 4 published · ↑ 2.3× vs 4-wk avg
Rank 3 ML / AI methods
Multimodal deep learning models benchmarked across omics, imaging and survival tasks
Recent preprints and publications test how multimodal models are benchmarked against baselines, metrics and validation designs for biological prediction.
6 preprints · 2 published · ↑ 2.1× vs 4-wk avg
Rank 4 ML / AI methods
Benchmarks test whether pretrained biological embeddings improve over simpler baselines
Recent preprints question whether pretrained biological embeddings and common evaluation metrics hold up under stricter downstream benchmarking.
10 preprints · 1 published · ↑ 2.0× vs 4-wk avg
Rank 5 Genomics
Integrating xQTL and diverse-ancestry GWAS with variant effect prediction
Calibrated variant effect data and ancestry-aware predictor evaluation are reshaping how variants and association hits are linked to genes and drug targets.
3 preprints · 4 published · ↑ 2.0× vs 4-wk avg
Rank 6 Genome-wide STRs and window-based SNP methods refine population inference across species↑ 1.7×GenomicsRank 7 Information and dynamical formalisms link biomarkers, cell states, and regulatory demand↑ 1.9×Systems biologyRank 8 Fusing heterogeneous inputs to predict drug, target and protein interactions↑ 1.7×ML / AI methodsRank 9 Computational RNA folding and interaction models extend beyond canonical base pairing↑ 1.7×Structural bio & protein designRank 10 Large cancer genome datasets assign mechanistic meaning to somatic alterations↑ 1.6×GenomicsRank 11 Learned representations of proteins, peptides and ligands drive binding prediction↑ 1.6×Structural bio & protein designRank 12 Inferring pathogen identity and variant composition from indirect proxy data↑ 1.5×GenomicsRank 13 Inference encoding biological structure across single-cell, spatial and plasma tumor dataSingle-cellRank 14 Open tools integrate annotation, motif, and metabolite data into interoperable resourcesTools & infrastructureRank 15 Community curation and multi-species annotation refine eukaryotic gene model resources↑ 1.5×GenomicsRank 16 Explicit error models and learned components drive identification in omics workflowsProteomicsRank 17 Read-level pipelines infer resistance and pathogen identity from raw sequencing readsGenomicsRank 18 Structure-encoding models classify enzymes, viruses, microbiome and barcode sequencesML / AI methodsRank 19 Immune variation and simulation frameworks couple population genetics with HLA biologyTools & infrastructureRank 20 Open software tools extract quantitative dynamics from live and spatial microscopy dataTools & infrastructure
Show 20 moreShow less
Rank 21 Circadian-gated slow-fast models infer time-dependent transitions in biological statesSystems biologyRank 22 ML prioritizes or designs candidate modulators for GPCR and TDP-43 aggregation targetsML / AI methodsRank 23 Time-varying directed and scale-dependent measures of neural interaction structureSystems biologyRank 24 Integrating experimental restraints and simulation to refine protein structure modelsStructural bio & protein designRank 25 Causal and network inference nominate disease targets from existing omic and network dataSystems biologyRank 26 Cryo-EM and cryo-ET density constraining generative atomic structure modelingStructural bio & protein designRank 27 LLM agents and knowledge graphs turn biomedical sources into auditable claimsSingle-cellRank 28 Learned and simulation-derived representations infer biomolecular structure and dynamicsStructural bio & protein designRank 29 Model-based inference of host, gene, and community traits from metagenomic sequencingSystems biologyRank 30 Learning pipelines couple fast RNA folding to codon design and structure predictionStructural bio & protein designRank 31 Symmetry- and geometry-aware learning models for molecular design and interpretationML / AI methodsRank 32 Machine learning on clinical brain data for dementia, epilepsy and MEG modelingML / AI methodsRank 33 Affine-gap and seed-based methods speed up sequence-to-graph and pangenome alignmentTools & infrastructureRank 34 Single-cell models forecast expression dynamics and predict nascent transcriptionSingle-cellRank 35 Comparative genomics links gene loss and lineage-specific evolution to distinct traitsGenomicsRank 36 Compressing and tokenizing DNA sequences for long-context genomic language modelsML / AI methodsRank 37 Reusable software for alignment, regulatory modelling and knockout phenotype networksTools & infrastructureRank 38 Junction and nuclear mechanics linked to EMT states and epithelial stratificationSystems biologyRank 39 Mass-offset and semi-tryptic searches widen peptide-level proteomic identificationProteomicsRank 40 Mapping noncoding GWAS variants to genes, splicing, and chromatin contactsEpigenomics