Top 40
View all ›PARTIAL Some sources or summaries were unavailable for this run.
BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Structural bio & protein design
Pretrained protein encoders enable residue-level localization and chirality transfer
Protein language model embeddings are now probed mechanistically and reused for residue-level binding and peptide design tasks.
15 preprints · 3 published · ↑ 4.7× vs 4-wk avg
Rank 2 Transcriptomics
Deep learning models are extending spatial and single-cell transcriptomic inference
Deep learning is now applied across spatial, single-cell and cell-free DNA transcriptomics, adding new supervision and imputation strategies.
15 preprints · 6 published · ↑ 2.9× vs 4-wk avg
Rank 3 Genomics
New methods infer genotypes, phasing, and population history from sequencing data
Pooled, sparse, and hybrid long- and short-read datasets are driving demand for inference tools that scale in runtime and accuracy.
6 preprints · 4 published · ↑ 2.9× vs 4-wk avg
Rank 4 Genomics
Integrating molQTL, diverse-ancestry GWAS and causal inference to prioritize drug targets
Combining xQTL, diverse-ancestry GWAS and causal inference now bridges noncoding variants to gene-level therapeutic target prioritization.
4 preprints · 6 published · ↑ 2.7× vs 4-wk avg
Rank 5 Tools & infrastructure
Direct inference of pathogen resistance from reads and non-sequencing proxies
End-user pipelines now detect resistance from raw nanopore and metagenomic reads, while MALDI-TOF spectra are tested as cheaper outbreak proxies.
3 preprints · 2 published · ↑ 2.1× vs 4-wk avg
Rank 6 ML predicts CRISPR off-targets, enzyme promiscuity, and peptide cyclization outcomes↑ 1.6×ML / AI methodsRank 7 Comparative genomics ties gene evolution and loss to lineage-specific traits↑ 1.6×GenomicsRank 8 Fusing heterogeneous side information to predict drug-drug and drug-target interactionsML / AI methodsRank 9 Metrics and validation designs challenge claimed gains in single-cell prediction↑ 1.5×Single-cellRank 10 Symmetry-aware graph learning for molecular generation and property predictionML / AI methodsRank 11 Genome-scale characterization of Leishmania, Plasmodium and alphavirus pathogensGenomicsRank 12 Generative models conditioned on cryo-EM and cryo-ET density recover atomic structuresStructural bio & protein designRank 13 Statistical controls and long-read isoform methods reshape expression analysisTranscriptomicsRank 14 Physics models and learned embeddings probe protein assembly, flexibility, and functionStructural bio & protein designRank 15 Learned models and free-energy workflows are accelerating biomolecular simulationStructural bio & protein designRank 16 Read-level analyses resolve viruses, plasmid hosts, and CRISPR arrays in metagenomesGenomicsRank 17 Machine learning models decode microbial genomes, enzymes, and microbiome featuresML / AI methodsRank 18 Mechanistic models link regulatory dynamics to cell states and signaling phenotypesSingle-cellRank 19 Simulations and designed proteins map energetic control of GPCR and channel functionStructural bio & protein designRank 20 Faster exact and graph-based alignment and placement algorithms for genomic dataTools & infrastructure
Show 20 moreShow less
Rank 21 Open-source tools unify heterogeneous biochemical and metagenomic data for queryingTools & infrastructureRank 22 Learned models infer circadian phase, EMT trajectories and gene expression dynamicsTranscriptomicsRank 23 Energy and metabolic costs shape neural coding, plasticity, and transcriptomic scalingML / AI methodsRank 24 Large cancer genome datasets assign mechanistic meaning to somatic alterationsGenomicsRank 25 Near-linear RNA folding extends to pseudoknots, chimeric evidence and codon designStructural bio & protein designRank 26 Probabilistic and learned models drive MS identification and annotation workflowsMetabolomicsRank 27 Computational reconstruction of transposable element activity and genome architectureGenomicsRank 28 Linking noncoding GWAS variants to target genes, splicing, and chromatin contactsEpigenomicsRank 29 Omics tools encode structure via tensor factorisation, enrichment and batch evaluationTools & infrastructureRank 30 Learned models for brain MRI, iEEG and SEEG tied to anatomy and expert labelsML / AI methodsRank 31 Computational models link local cell rules to tissue-scale invasion and polaritySystems biologyRank 32 Open in silico metabolite predictors benchmarked against human radiolabeled ADME dataTools & infrastructureRank 33 LLM agents grounded in raw single-cell data for annotation and perturbation predictionSingle-cellRank 34 Learned models predict infection potential, peptide retention, and cancer diagnosisTools & infrastructureRank 35 Pretrained foundation models span medical images, MEG signals, and molecular SMILESML / AI methodsRank 36 Structured knowledge graphs and ontologies for standardized biomedical knowledgeTools & infrastructureRank 37 Richness estimators predict undiscovered phage species across eight bacterial hostsGenomicsRank 38 Structure tools expose hidden conformations; zero-shot models fail under epistasisStructural bio & protein designRank 39 Computational prioritization of candidates ahead of experimental testingStructural bio & protein designRank 40 Automated extraction of cell dynamics from time-resolved microscopy imagesTools & infrastructure