Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Single-cell
Learned models infer cell-state dynamics and gene regulation from single-cell profiles
Single-cell and spatial datasets now reach tens of millions of cells, letting learned models be trained at atlas scale.
8 preprints · 3 published · ↑ 2.8× vs 4-wk avg
Rank 2 Tools & infrastructure
Reference-aware tools automate chromosome-scale assembly and haplogroup inference
Late-2025 and early-2026 releases pair reference-aware methods with containerized, QC-integrated pipelines for assembly, haplogroup and contamination workflows.
2 preprints · 9 published · ↑ 2.5× vs 4-wk avg
Rank 3 ML / AI methods
Joint multimodal models couple heterogeneous biological data within single architectures
Recent preprints and publications couple brain imaging, DNA methylation and drug-property targets within single learning models.
3 preprints · 6 published · ↑ 2.4× vs 4-wk avg
Rank 4 Genomics
Multimodal and cell-type functional data integrated into variant and cancer inference
Single-cell and cfDNA datasets now supply cell-type and molecular features that refine rare-variant, germline and ctDNA inference.
4 preprints · 6 published · ↑ 2.1× vs 4-wk avg
Rank 5 ML / AI methods
Graph learning on molecular and biological networks predicts drug and protein links
Recent papers combine attention, contrastive, and adversarial designs with heterogeneous biomedical graphs for prediction and interpretability.
3 preprints · 9 published · ↑ 2.0× vs 4-wk avg
Rank 6 Multiscale computation links sequence, conformational dynamics and phase behavior↑ 1.9×Structural bio & protein designRank 7 Computational methods infer spatial and single-cell structure beyond direct measurement↑ 1.7×Single-cellRank 8 LLMs adapted and benchmarked against curated biomedical ground truth↑ 1.8×ML / AI methodsRank 9 Complete and pangenome assemblies illuminate genome architecture across plants and insects↑ 1.8×GenomicsRank 10 LLM agents tied to verifiable references for biological tools, code, and figures↑ 1.7×Tools & infrastructureRank 11 Attention-based encoders predict peptide, protein, and CRISPR off-target outcomes↑ 1.6×ML / AI methodsRank 12 Benchmarking omics pipelines and explainable models for biomarker and cell-type discovery↑ 1.5×ML / AI methodsRank 13 Long-read methods expose errors and ambiguity in mixed and repetitive genomesGenomicsRank 14 Network priors and rank statistics guide omics inference for disease genes and tissuesSystems biologyRank 15 Simulation and coalescent tools generalize to multichromosome and ecological models↑ 1.5×Tools & infrastructureRank 16 Slow regulatory variables and noise reshape accessible single-cell state trajectoriesSystems biologyRank 17 Packaged tools for multi-omics, microbiome, network and single-cell data analysisTools & infrastructureRank 18 Generative and predictive models are steering de novo protein and peptide designStructural bio & protein designRank 19 Computational tools map Plasmodium population genomics and recurrence inferenceTools & infrastructureRank 20 Targeted pipelines wrap learned components for specific protein prediction tasksTools & infrastructure
Show 19 more
Rank 21 Protein language models guide antibody, peptide and epitope design and predictionML / AI methodsRank 22 Deep models map nanopore, sequence, and expression data to biomolecular predictionsML / AI methodsRank 23 Machine learning and template methods extend RNA structure and design tasksStructural bio & protein designRank 24 Open-source software for simulating lipid, gene circuit, and regulatory sequence modelsTools & infrastructureRank 25 Diverse-ancestry and mouse-strain profiling links variants to expression and traitsGenomicsRank 26 Language and graph foundation models target drug repurposing and molecule generationML / AI methodsRank 27 Reproducible analysis tools coupled to curated, provenance-tracked biological dataTools & infrastructureRank 28 Statistical methods target methylation dysregulation and censored biomarker measurementsEpigenomicsRank 29 Computable models link cell migration, polarity, and signaling across cell scalesSystems biologyRank 30 Computational tools link molecular features to outcomes, cell identity, and genotypeSingle-cellRank 31 Reconstructing ancient plant genome duplications and protein domain evolutionGenomicsRank 32 Single-cell transcriptomes resolve cell-type-specific isoform and splicing programsSingle-cellRank 33 Protein-ligand binding models assessed under prospective and setting-specific conditionsStructural bio & protein designRank 34 Non-genetic molecular memory and generative design of transcriptional activatorsSystems biologyRank 35 Learned models reshape cryo-EM reconstruction and electron microscopy acquisitionStructural bio & protein designRank 36 Models link behavioral feedback, emulation and spatial shedding to infection outcomesSystems biologyRank 37 Biological priors and domain adaptation sharpen cell-type inference under scarce dataSingle-cellRank 38 Genome-scale sequence evolution and recombination-aware gene construct designGenomicsRank 39 Recovering kinetic and circadian timing from untimed or nascent transcriptomic readoutsSystems biology