Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Single-cell
Transport-constrained encoders model single-cell dynamics and cross-modal expression
Unpaired snapshots, perturbations, and spatial modalities now demand generative models that link latent cell states across conditions.
15 preprints · 3 published · ↑ 3.8× vs 4-wk avg
Rank 2 ML / AI methods
Graph neural networks predict drug, protein and host-pathogen interactions
Graph-based deep learning now covers host-pathogen, synergy, CNS penetration and allergenicity prediction, with explainability and debiased benchmarking.
6 preprints · 6 published · ↑ 3.3× vs 4-wk avg
Rank 3 Single-cell
Methods separate measurement noise from biology in single-cell and spatial RNA data
Recent preprints and publications extend imputation, denoising, deconvolution, and spatial mapping of scRNA-seq and spatial transcriptomics data.
8 preprints · 7 published · ↑ 3.3× vs 4-wk avg
Rank 4 Structural bio & protein design
Sequence-based deep learning predicts protein binding and interaction properties
Protein language model embeddings and attention architectures are now applied to binding affinity, interaction, and AlphaFold3 confidence prediction.
9 preprints · 2 published · ↑ 3.1× vs 4-wk avg
Rank 5 ML / AI methods
Protein language models now drive antimicrobial discovery and sequence redesign
Learned protein representations now span metagenomic-scale peptide discovery and generative redesign conditioned on structural or evolutionary context.
4 preprints · 4 published · ↑ 2.2× vs 4-wk avg
Rank 6 Sampling Frames Determine Which Microbial Lineages Enter Genomic Analyses↑ 2.1×GenomicsRank 7 Learned models of neural connectivity and latent dynamics across EEG, SEEG, and ECoG↑ 2.1×ML / AI methodsRank 8 Deep learning models predict noncoding regulatory activity to prioritize variants↑ 2.0×GenomicsRank 9 Quantitative models infer cell-state dynamics from single-cell and perturbation data↑ 1.9×Single-cellRank 10 Comparative genomics links mobile elements and effectors to host and ecological context↑ 1.7×GenomicsRank 11 Sequence and nucleosome context jointly govern chromatin structure and accessibility↑ 1.9×EpigenomicsRank 12 Population genomes resolve domestication history, diversity and inherited methylation↑ 1.8×GenomicsRank 13 Inference from near-raw measurements across noise, counting, and classification↑ 1.7×ML / AI methodsRank 14 Constraint-based and multi-omic models map disease-specific metabolic regulation↑ 1.7×Systems biologyRank 15 Deep learning models predict regulatory activity and 3D chromatin from DNA sequence↑ 1.6×ML / AI methodsRank 16 Gapless T2T and population multiomic references anchor downstream genome analysis↑ 1.6×GenomicsRank 17 Haplotype-resolved cattle genome and tools for Y-haplogroup calling and assembly QC↑ 1.6×GenomicsRank 18 GPU and SIMD acceleration of sequence alignment, motif detection, and read simulation↑ 1.5×Tools & infrastructureRank 19 Statistical gating and benchmarking embedded in open omics analysis tools↑ 1.6×TranscriptomicsRank 20 Deep models predict HIV diagnoses, protein succinylation, and RNA methylation sites↑ 1.5×ML / AI methods
Show 19 more
Rank 21 Learned models and benchmarks refine genomic variant calling and interpretationGenomicsRank 22 Membrane oscillators and ion transport emerge from local kinetic and physical rulesSystems biologyRank 23 Causal and enrichment criteria qualify network-derived target predictionsSystems biologyRank 24 Benchmarks and new methods for extracting structure from transcriptomic dataML / AI methodsRank 25 Network models of gene transfer and phylogenetic diversity in microbial sequence dataGenomicsRank 26 Quantifying conformational dynamics across MD, coarse-grained and predicted ensemblesStructural bio & protein designRank 27 Symbolic, thermodynamic and kinetic models of reaction network dynamicsSystems biologyRank 28 Energetic and dynamical assessment of protein structures and binding ensemblesStructural bio & protein designRank 29 Lipid transfer gating in ATG2A and pore-mediated endosomal escape of LNPsStructural bio & protein designRank 30 Protein language model embeddings are being packaged into prediction and design toolsTools & infrastructureRank 31 LLMs and networks infer disease, gene, and therapeutic relationships from dataSystems biologyRank 32 Open-source workflows for information-theoretic and multi-omics analysis of biological…Tools & infrastructureRank 33 Open-source Python tools pair mechanistic models with ML and Bayesian calibrationSystems biologyRank 34 New tools predict G-quadruplexes, Z-DNA and long-read transcript isoformsML / AI methodsRank 35 Generative models now condition on explicit pocket, lead, prior, or reward informationStructural bio & protein designRank 36 Learning rules and circuit architectures fit to reaction-time and policy behaviorSystems biologyRank 37 Open-source LLM agents map natural-language queries onto genomic and model databasesTools & infrastructureRank 38 Demographic noise, dispersal, and nutrient limits reshape predator-prey cyclesSystems biologyRank 39 Minimizer density modeled via inverse spacing; multiminimizers approach known limitTools & infrastructure