Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
ML / AI methods
Fusing graph, language-model and structure embeddings for biomolecular prediction
Joint modelling of heterogeneous biomolecular representations now spans drug-target, drug-drug, drug-gene, solubility and antibody prediction tasks in this…
6 preprints · 10 published · ↑ 4.2× vs 4-wk avg
Rank 2 Genomics
Algorithmic reformulation of post-contig long-read assembly: scaffolding and gap filling
Published and preprint methods now target the fragmentation that persists after initial long-read contig construction.
4 preprints · 4 published · ↑ 2.3× vs 4-wk avg
Rank 3 ML / AI methods
Language models predict TCR-peptide recognition and protein interaction partners
Protein language models are now being applied to TCR-antigen and protein-interaction prediction, including unseen peptides and MHC.
1 preprint · 6 published · ↑ 2.1× vs 4-wk avg
Rank 4 Systems biology
Mechanistic simulation with inferred contact structure for sparse transmission data
Sparse observations make simulators that reconstruct contact structure useful for PrEP allocation, H5N1 risk, and respiratory virus surveillance.
2 preprints · 1 published · ↑ 2.0× vs 4-wk avg
Rank 5 Epigenomics
Sequence-based models and cohesin resources map regulatory sites and TF binding
Recent preprints and published work extend regulatory site catalogs, sequence-to-function models, and silencer atlases across TF binding and chromatin context.
3 preprints · 6 published · ↑ 1.7× vs 4-wk avg
Rank 6 Learned generative and surrogate models span antibodies, transport, and biomechanics↑ 1.9×ML / AI methodsRank 7 Computational prediction of protein-ligand binding from predicted and docked structures↑ 1.7×Structural bio & protein designRank 8 Deep learning models predict biomolecular interactions from sequence and structure↑ 1.5×ML / AI methodsRank 9 Biological networks embedded in graph learning for multi-omics and regulatory inference↑ 1.5×ML / AI methodsRank 10 Multi-ancestry and rare-variant extensions of genetic association and fine-mapping↑ 1.7×GenomicsRank 11 Atlas-scale single-cell and spatial resources standardize annotation and deconvolution↑ 1.6×TranscriptomicsRank 12 Curated dietary, microbial and reaction inputs linked to metabolic model endpoints↑ 1.6×Tools & infrastructureRank 13 Algorithms for RNA-seq strandedness, fusion calling and nanopore modification detectionTranscriptomicsRank 14 Learned models of tandem mass spectra drive peptide identification and quantification↑ 1.5×ProteomicsRank 15 Computational inference of spatial organization and cell identity from spatial omicsTranscriptomicsRank 16 Scale and pretraining do not reliably improve single-cell foundation model performanceSingle-cellRank 17 Protonation states gate conformational transitions in channels, enzymes and condensates↑ 1.6×Structural bio & protein designRank 18 Open-source R and Julia tools decompose and integrate multiomics and spatial dataTools & infrastructureRank 19 Causal and mediation estimators extended to high-dimensional, nonlinear omics dataML / AI methodsRank 20 Machine learning models link single-cell states to drug and immunotherapy responseSingle-cell
Show 20 more
Rank 21 Generative models are conditioned on pockets, 3D structure, and explicit latent spacesStructural bio & protein designRank 22 Computational platforms map drug-target networks, ADMET, and peptide candidatesML / AI methodsRank 23 Computational biomarkers for neurodegeneration from MRI, genetics, and methylation ageEpigenomicsRank 24 Supervised learning links sequence and variant data to protein function and stabilityML / AI methodsRank 25 Curated RNA resources and nanopore methods map circRNA modifications and interactionsTools & infrastructureRank 26 Deep learning predicts interface quality, RNA flexibility and disordered-region contactsStructural bio & protein designRank 27 TED domain profile HMMs from AlphaFold models enable remote homology detectionStructural bio & protein designRank 28 Computational reconstruction of 3D chromosome structures from Hi-C contact dataEpigenomicsRank 29 Simulations probe filament design, condensin gating and confined nucleic acid foldingStructural bio & protein designRank 30 Monoculture metabolic profiles and interaction models predict microbial community behaviorSystems biologyRank 31 Benchmarking omics tools for sensitivity, robustness, and ground-truth accuracySystems biologyRank 32 Hyperspectral latent factors and deep learning extend plant genomic predictionGenomicsRank 33 Markov state models map conformational landscapes in drug resistance and allosteryStructural bio & protein designRank 34 Genomic methods infer relatedness, selection, and ancestry without pedigree dataGenomicsRank 35 LLM agents and traceable toolkits are automating bioinformatics analysis workflowsTools & infrastructureRank 36 Spatial 3D mapping resolves regulatory networks in gastrulation, limb bud and fish heartSingle-cellRank 37 Statistical recovery of expression lost to dropout and unresolved cell proportionsSingle-cellRank 38 Percolation and energy landscape frameworks model brain dynamical statesSystems biologyRank 39 New tools and resources reconcile genome comparison and repeat annotation across…Tools & infrastructureRank 40 Refining proteomics searches and genome annotation for PTMs and selenoproteinsProteomics