Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
ML / AI methods
Graphs and protein language models jointly encode structure and sequence for prediction
Recent preprints and papers pair protein language model embeddings with graph or sparse-attention encodings of structure and interactions.
7 preprints · 5 published · ↑ 4.0× vs 4-wk avg
Rank 2 Tools & infrastructure
Open workflows for reprocessed and integrated omics data analysis
Public mass-spectrometry archives and single-cell and spatial datasets are increasingly reanalyzed through open, reproducible computational pipelines.
3 preprints · 7 published · ↑ 3.1× vs 4-wk avg
Rank 3 Tools & infrastructure
Nanopore direct RNA methods advance basecalling, isoform annotation, and RBP binding maps
Long-read direct RNA sequencing now supports improved basecalling, tRNA and isoform annotation, and transcript-resolved RBP binding maps.
1 preprint · 8 published · ↑ 3.0× vs 4-wk avg
Rank 4 Structural bio & protein design
Learned predictors steer generative design of binders, enzymes and small molecules
Predictive objectives are increasingly embedded in generative and search pipelines across ligands, RNA, enzymes, and antibodies.
9 preprints · 6 published · ↑ 2.9× vs 4-wk avg
Rank 5 ML / AI methods
Learned models integrate multi-omic profiles for biomarker and phenotype prediction
January 2026 reports apply self-supervised, contrastive and attention-based learning to purify, integrate and rank multi-omic features.
4 preprints · 5 published · ↑ 2.8× vs 4-wk avg
Rank 6 Deep learning couples H&E histology with spatial transcriptomics and multiplex imaging↑ 2.7×ML / AI methodsRank 7 Benchmarking and divergence metrics reconcile bulk and single-cell transcriptomic data↑ 2.3×Single-cellRank 8 Reprocessed and benchmarked proteome resources expose evidence behind protein detection↑ 2.3×ProteomicsRank 9 Inferring kinetic parameters from stochastic biochemical and single-molecule dynamics↑ 2.3×Structural bio & protein designRank 10 Benchmarks and models probe protein conformational ensembles beyond static structures↑ 2.2×Structural bio & protein designRank 11 Open deep learning pipelines extract plant root, fruit, and growth traits from images↑ 2.3×Tools & infrastructureRank 12 Deep learning and benchmarking reshape nanopore and single-cell sequencing analysis↑ 1.9×GenomicsRank 13 2D sections can distort 3D tissue patterns; spatial structure is now modeled directly↑ 1.8×Tools & infrastructureRank 14 Mutation and dependency data combined to prioritise cancer targets and progression models↑ 1.8×Tools & infrastructureRank 15 Single-cell and genetic maps link disease risk to cell-type regulatory programs↑ 1.8×Single-cellRank 16 Single-model joint inference of latent, censored, and cross-site structure in omics data↑ 1.8×ML / AI methodsRank 17 Explicit admissibility criteria for biochemical models in parameter and state spaces↑ 1.8×Systems biologyRank 18 Deep models condition genomic predictions on tissue, clinical, and regulatory context↑ 1.7×ML / AI methodsRank 19 Machine-readable metadata and identifiers make biological data validatable and reusable↑ 1.7×Tools & infrastructureRank 20 Simulation and forecasting of viral evolution under host and drug selection pressure↑ 1.6×Systems biology
Show 20 more
Rank 21 Assembly-based pangenomes link genome structure to adaptation and agronomic traits↑ 1.6×GenomicsRank 22 Ligands and ionic conditions reshape conformational free-energy landscapes↑ 1.6×Structural bio & protein designRank 23 Retrieval grounding checks biomedical LLM outputs against external knowledge sources↑ 1.5×Tools & infrastructureRank 24 Compressed indexes and co-designed accelerators for scalable sequencing data analysisTools & infrastructureRank 25 Reference standards and callers expose false positives in variant detectionGenomicsRank 26 MD and enhanced sampling resolve ligand, molecular glue, and allosteric binding mechanismsStructural bio & protein designRank 27 Predicting and generating binders for antibodies, TCRs, peptides, and molecular gluesStructural bio & protein designRank 28 Deep learning segmentation quantifies embryo, organoid and osteocyte imaging dataML / AI methodsRank 29 Interaction structure and environmental stress reshape competitive population dynamicsSystems biologyRank 30 Reference-quality bird genome assemblies expand taxonomic and chromosomal coverageGenomicsRank 31 Learned classifiers applied to cfDNA fragments, metabolomics, and resistance profilesML / AI methodsRank 32 Molecular representation learning and flow matching drive property-guided molecule designML / AI methodsRank 33 Network structure and spatial heterogeneity reshape epidemic transmission dynamicsSystems biologyRank 34 Structure-constrained learned search guides RNA, regulatory, and enzyme sequence designStructural bio & protein designRank 35 Multiplexed proteomics stratifies disease states using protein panels and PTM readoutsProteomicsRank 36 Transposable elements supply new miRNA target sites and alternative exonsTranscriptomicsRank 37 Structural variants and polygenic prediction refine genetic association inferenceGenomicsRank 38 Transformer and point-cloud networks refine cryo-EM maps and conformational heterogeneityStructural bio & protein designRank 39 Chromosome-level assemblies link genome structure to insect and orchid adaptationGenomicsRank 40 Benchmarks and open tools for inferring genomic variables from sequencing readsGenomics