Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Tools & infrastructure
Nanopore and short-read tools for microbial typing, methylation and variant detection
Recent preprints and one published suite pair typing, variant and QC analysis with direct benchmarks against established microbial genomics methods.
6 preprints · 1 published · ↑ 2.2× vs 4-wk avg
Rank 2 ML / AI methods
Fused learned representations drive prediction across protein, RNA, and molecular tasks
Recent preprints and papers pair learned embeddings with fusion architectures and new benchmarks for protein, RNA, and molecular prediction.
5 preprints · 7 published · ↑ 2.1× vs 4-wk avg
Rank 3 Tools & infrastructure
Long-read tools quantify somatic mosaicism and repeat variation in human genomes
Recent preprints extend long-read and repeat-aware tools to somatic mosaicism at HTT repeats and across structural and small variants.
7 preprints · ↑ 2.1× vs 4-wk avg
Rank 4 Genomics
Pangenomes and orthogonal evidence layers refine eukaryotic gene and variant calls
Large T2T-level assemblies and pangenomes now pair with mass spectrometry and deep learning to link genome structure with trait and variant effects.
2 preprints · 1 published · ↑ 2.0× vs 4-wk avg
Rank 5 Single-cell
Computational methods recover structure from sparse, noisy transcriptomic measurements
A 30-dataset benchmark finds traditional scRNA-seq imputation often outperforms deep learning approaches.
8 preprints · 3 published · ↑ 1.6× vs 4-wk avg
Rank 6 Frequency-dependent selection and long-term evolution shape bacterial lineage structure↑ 1.6×Systems biologyRank 7 Pretrained protein language models extended to structure, binding, and model selection↑ 1.6×ML / AI methodsRank 8 Benchmarking variant annotators and polygenic scores for genotype-to-phenotype inference↑ 1.6×GenomicsRank 9 Deep learning models predict chromatin looping and 3D genome contacts↑ 1.5×EpigenomicsRank 10 Generative and ML models propose protein sequences that are then scored and filteredStructural bio & protein designRank 11 Genomic language models are predicting viral host range, pathogenicity and taxonomyML / AI methodsRank 12 Flow and diffusion generators sample molecules, RNA structures, and MD trajectories↑ 1.6×ML / AI methodsRank 13 Pretrained protein language model embeddings paired with structural and graph contextStructural bio & protein designRank 14 Transformer and unsupervised models are tested on high-dimensional biological dataML / AI methodsRank 15 Faster flux consistency testing for context-specific metabolic model reconstructionML / AI methodsRank 16 Benchmarking MD-coupled docking and solvent-aware workflows for structure-based designStructural bio & protein designRank 17 Learned sequence models predict RNA structure, miRNA targeting and protein-RNA bindingML / AI methodsRank 18 Metagenomics uncovers novel soil microbes and large DNA viruses in environmental samplesGenomicsRank 19 Long-read assemblies and reference choice reshape isoform-level genetic and expression…TranscriptomicsRank 20 Cell-type-resolved genetic regulation of gene expression in brain diseaseGenomics
Show 20 more
Rank 21 Reduced-resolution models extend biomolecular simulation timescales and system sizesStructural bio & protein designRank 22 Sequence language models move from protein residues to mRNA nucleotide designML / AI methodsRank 23 Integrative and force-aware models capture flexible RNA and RNA-binding ensemblesStructural bio & protein designRank 24 Atomistic simulations trace remote binding to allosteric control of protein conformationsStructural bio & protein designRank 25 Optimal transport infers brain routing and cellular stochastic dynamicsSystems biologyRank 26 Linking tissue-scale myelin and field effects to network-level brain functionSystems biologyRank 27 Learned sequence models steered toward several design objectives at onceTools & infrastructureRank 28 Multi-layer profiling resolves transcription factor programs in distinct cell statesSystems biologyRank 29 Modular software platforms advance site-specific glycoproteomics data analysisProteomicsRank 30 Long-read and bisulfite-sequencing tools resolve DNA and RNA methylation statesEpigenomicsRank 31 Protein-level sequence and structure predict phage-host links and defense domain evolutionGenomicsRank 32 Computational models resolve cell-cell communication and drug response in tissue dataSingle-cellRank 33 Molecular signatures define neurodegeneration subtypes and candidate biomarkersProteomicsRank 34 Shared network models generate both hippocampal place and time cell representationsSystems biologyRank 35 New tools for plant contig scaffolding, repeat detection and metabolite reference dataTools & infrastructureRank 36 Spatial organization of mitochondria, neuropeptides and cell types in neural tissueSystems biologyRank 37 Ensemble models resolve nucleosome-scale chromatin organization in single cellsEpigenomicsRank 38 Software tools correcting and accelerating raw sequencing data processingTools & infrastructureRank 39 Simulations couple molecular binding to emergent cell and tissue mechanicsSystems biologyRank 40 Space-time trade-offs in k-mer filters, superstrings, and repertoire searchTools & infrastructure