Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
ML / AI methods
Machine learning models predict drug response and treatment failure in cancer
Recent preprints and papers apply interpretable, foundation-model, and benchmarking approaches to tumor and single-cell data for therapy outcome prediction.
5 preprints · 5 published · ↑ 2.9× vs 4-wk avg
Rank 2 Transcriptomics
Computational models infer spatial gene activity from tissue imaging and transcriptomics
Recent preprints and publications extend spatial transcriptomics models toward causal inference, batch-bias detection, foundation models, and 3D augmentation.
9 preprints · 3 published · ↑ 2.7× vs 4-wk avg
Rank 3 Structural bio & protein design
Graph models and open tools for mining protein interfaces and glycan 3D structures
Recent preprints and a published package apply graph representations and atomic-level features to interface and glycan structure mining.
3 preprints · 1 published · ↑ 2.3× vs 4-wk avg
Rank 4 Tools & infrastructure
Simulation against known truth benchmarks sequencing and epigenomic analysis software
Controlled simulation with known ground truth now quantifies mapping, variant calling, and modification detection tool performance on long-read and short-read…
11 preprints · 1 published · ↑ 2.1× vs 4-wk avg
Rank 5 Single-cell
Single-cell data attribute genetic and perturbation effects to specific cell types
Airqtl, SIDISH and iPS2-seq move eQTL, risk and perturbation analyses from bulk averages to cell-state resolution.
6 preprints · 7 published · ↑ 2.1× vs 4-wk avg
Rank 6 Biology-constrained embeddings for cell typing, regulons, and multi-omic integration↑ 2.0×Single-cellRank 7 Computational frameworks link gut microbial metabolism to host phenotypes and diet↑ 1.8×Systems biologyRank 8 Residue-level energetics and networks resolve protein conformational change and binding↑ 2.0×Structural bio & protein designRank 9 Chromosome-scale assemblies link genome architecture to size and demographic history↑ 1.9×GenomicsRank 10 Coarse-grained and enhanced-sampling models generate conformational state ensembles↑ 1.8×Structural bio & protein designRank 11 RNA-binding proteins link exon usage, chromatin marks, and mRNA decay regulation↑ 1.6×EpigenomicsRank 12 Explicit computational frameworks map and predict neural responses across modalitiesML / AI methodsRank 13 Chemical language models extend to 3D generation, toxicity, and drug-target predictionML / AI methodsRank 14 Post-calling tools integrate population, phenotypic, and database evidence for variantsGenomicsRank 15 Compact k-mer and minimizer structures are reused for pangenome alignment and sketchingGenomicsRank 16 Genome-scale models and comparative genomics map bacterial metabolic diversityGenomicsRank 17 Untargeted MS workflows extend peak picking, drug exposure and dereplicationMetabolomicsRank 18 Standardized benchmarking of transcriptomic gene sets and isoform features across datasetsSingle-cellRank 19 Foundation models and knowledge-guided learning advance cancer tissue image analysisML / AI methodsRank 20 Genome-scale criteria are redrawing bacterial taxonomic boundaries and HGT detectionGenomics
Show 20 more
Rank 21 Computational Tracing of Atoms, Stereochemistry and Assembly in Metabolite ChemistryMetabolomicsRank 22 Mapping GWAS loci to genes and cell types across psychiatric and migraine studiesGenomicsRank 23 Software defines haplotype blocks and crossovers from genotype data for association workTools & infrastructureRank 24 Physically grounded evaluation of learned samplers for protein conformational ensemblesStructural bio & protein designRank 25 Self-supervised and autoencoder models learn biological representations from varied dataML / AI methodsRank 26 Predictive neoantigen prioritization for vaccines, response scoring and off-target riskGenomicsRank 27 Learned models replace manual annotation in cell microscopy analysisML / AI methodsRank 28 Software tools standardise, benchmark and visualise derived omics and structural outputsTools & infrastructureRank 29 Learned embeddings and fusion models predict genomic, RNA and molecular associationsML / AI methodsRank 30 Benchmarked choices for GPCR free energy, residence time and fragment pre-screeningStructural bio & protein designRank 31 Predicted protein and RNA structures inform substitution, residue design, and energy…Structural bio & protein designRank 32 Generative augmentation and transformers address small-data peptide classificationML / AI methodsRank 33 Curated fine-tuning steers language models toward functional protein and plasmid designsStructural bio & protein designRank 34 Neural approximation of mechanistic reaction, metabolic, and single-cell dynamicsSystems biologyRank 35 Assay-specific preprocessing and interval tools speed long-read and single-cell genomicsTools & infrastructureRank 36 Structure-conditioned design now spans nuclease scaffolds and antibody repertoiresStructural bio & protein designRank 37 Curated references and coverage-aware models reduce false taxa in metagenomic profilingTools & infrastructureRank 38 New association methods partition polygenic, epistatic and regulatory genetic effectsSystems biologyRank 39 Long-read and single-cell tools probe chromosome architecture across speciesEpigenomicsRank 40 Genealogy-aware inference methods for lineages, population genetics and complex traitsGenomics