Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
ML / AI methods
Graph neural networks extend molecular interaction and property prediction across targets
Recent items pair graph embeddings and fingerprint fusion with compound, protein, and drug-combination prediction tasks across diverse endpoints.
2 preprints · 6 published · ↑ 2.9× vs 4-wk avg
Rank 2 Genomics
Learned sequence models score noncoding and rare variants for functional prioritization
Learned models now score regulatory activity and variant effects directly from sequence, feeding rare-variant and noncoding prioritization.
2 preprints · 6 published · ↑ 2.8× vs 4-wk avg
Rank 3 ML / AI methods
Learned models decode neural interactions across EEG, SEEG, ECoG and connectomes
Recent preprints and a shared-space ECoG study extend neural decoding toward subject alignment and model-brain correspondence.
6 preprints · 1 published · ↑ 2.7× vs 4-wk avg
Rank 4 Single-cell
Computational methods map and denoise single-cell and spatial transcriptomic data
Recent preprints and publications benchmark imputation, deconvolution, and cross-modal mapping tools across scRNA-seq and spatial transcriptomics platforms.
4 preprints · 8 published · ↑ 2.6× vs 4-wk avg
Rank 5 ML / AI methods
Plasmid and variant mining links mobile genetic content to hosts and function
Late-2025 items pair plasmid and genome inventories with benchmarks or ecological context, such as soil plasmidomes and Hi-C host linkage.
3 preprints · 4 published · ↑ 2.5× vs 4-wk avg
Rank 6 Learned noise and interaction structure from sparse high-dimensional biological data↑ 2.1×ML / AI methodsRank 7 Sampling frame and marker choice decide which microbial lineages enter genomic analysis↑ 2.3×GenomicsRank 8 Transport and graph constraints on encoders for perturbation and spatial prediction↑ 1.8×Single-cellRank 9 Omitted epistasis inflates GWAS nulls; multitrait and metabolite maps extend loci↑ 1.9×GenomicsRank 10 Sequence-based models predict protein interactions, RNA structure and binding affinity↑ 1.7×Structural bio & protein designRank 11 GPU and SIMD acceleration of alignment, simulation and metagenomic sequence workflows↑ 1.8×Tools & infrastructureRank 12 Temporally resolved single-cell data fitted to dynamic models infer cell fate behavior↑ 1.7×Single-cellRank 13 Gapless T2T and population-scale reference genomes anchor baseline genomic resources↑ 1.7×GenomicsRank 14 Multi-omic metabolic models infer disease-specific states across cancer and IBD↑ 1.6×Systems biologyRank 15 Deep models on structured inputs predict epidemiological and molecular interactions↑ 1.6×ML / AI methodsRank 16 Computational tools for Y-haplogroup calling, Z-DNA prediction, LD mapping and assembly QC↑ 1.6×GenomicsRank 17 Open-source packages make spatial, single-cell and multi-omics analyses reproducible↑ 1.6×Tools & infrastructureRank 18 DNA sequence and nucleosome organization shape chromatin structure and accessibility↑ 1.8×EpigenomicsRank 19 Epidemic simulators pair RL reward design and ABC calibration for transmission modelingSystems biologyRank 20 Language models condition protein and gene design on evolutionary and genomic context↑ 1.5×Structural bio & protein design
Show 19 moreShow less
Rank 21 Sequence-based deep learning predicts protein binding, interaction, and allergenicityStructural bio & protein designRank 22 Structure-based models link reaction network topology, parameters, and dynamical behaviorSystems biologyRank 23 Baseline-aware benchmarking of computational methods across omic data typesML / AI methodsRank 24 Dimensionality, flexibility, and NMR-restrained sampling for macromolecular motionsStructural bio & protein designRank 25 Network-based gene and module prioritization gains causal and validation criteriaSystems biologyRank 26 Hybrid mechanistic-ML tools and noise-aware inference for dynamical biological modelsML / AI methodsRank 27 DNA deep learning models predict 3D chromatin and cell type-specific gene expressionML / AI methodsRank 28 Small-molecule binding and oligomer dynamics in disordered and allosteric proteinsStructural bio & protein designRank 29 Topological tests and higher-order complexes expose structure in brain networksSystems biologyRank 30 Open pipelines automate protein binder design, docking, and coarse-grained topologiesStructural bio & protein designRank 31 Target-conditioned diffusion and flow models generate 3D ligands and therapeutic peptidesStructural bio & protein designRank 32 Encoded decision rules in sequencing tools for RNA-protein and splicing analysisTranscriptomicsRank 33 Multiscale simulations link cancer cell shape, membrane lipids and immune primingStructural bio & protein designRank 34 Comparative genomes reveal chromosome fusions, hybrid coupling, and WGD inference biasesGenomicsRank 35 Fitting learning rules and circuit architectures to behavior and spiking dynamicsSystems biologyRank 36 Open-source LLM agents map natural language to genomic and kinetic biology resourcesTools & infrastructureRank 37 Scalable graph algorithms for phylogenetic networks and pangenome bubble detectionGenomicsRank 38 Network size and frequency allocation govern oscillatory synchrony and resonanceSystems biologyRank 39 Noise, dispersal, and nutrient limits reshape predator-prey population dynamicsSystems biology