Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
ML / AI methods
Fusing language and sequence embeddings with graph architectures for drug prediction
Pretrained language and sequence embeddings are now being paired with graph and state-space encoders across drug, RNA and antimicrobial prediction tasks.
2 preprints · 8 published · ↑ 4.3× vs 4-wk avg
Rank 2 Genomics
Benchmarked genome annotation, assembly, and scaffolding tools across diverse datasets
Recent preprints and publications pair genome analysis methods with benchmarks on reference genomes, long reads, and multi-species datasets.
4 preprints · 3 published · ↑ 3.3× vs 4-wk avg
Rank 3 Single-cell
Computational methods for inferring networks and cell states from single-cell data
Sparse, noisy single-cell and spatial measurements motivate methods for cell-type-aware inference across transcriptomic and multiomic data.
9 preprints · 5 published · ↑ 3.1× vs 4-wk avg
Rank 4 ML / AI methods
Pretrained sequence language models repurposed for protein, peptide and RNA prediction
Pretrained sequence models now pair with controlled sampling, leakage-aware datasets and tokenization choices for downstream prediction.
3 preprints · 7 published · ↑ 2.9× vs 4-wk avg
Rank 5 Epigenomics
Computational models linking noncoding DNA, chromatin marks and gene regulation
Long-read methylation, single-cell multiomics and transformer models now predict chromatin marks and regulatory links from sequence data.
5 preprints · 4 published · ↑ 2.7× vs 4-wk avg
Rank 6 Structure-based models predict binding shifts from epitope edits and protein design↑ 2.7×Structural bio & protein designRank 7 Evaluated computational tools for glycoproteomics, transcriptomics, and structure search↑ 2.7×Structural bio & protein designRank 8 Reproducible workflows standardize post-acquisition imaging, omics, and assay data↑ 2.5×Tools & infrastructureRank 9 Reference genomes and cell-type taxonomies meet single-nucleus aging profiles in brain↑ 2.3×Single-cellRank 10 Pipelines address sequencing artefacts and defective genomes in viral and microbial data↑ 2.3×Tools & infrastructureRank 11 Reusing learned models across datasets, species and contexts when labels are scarce↑ 2.2×ML / AI methodsRank 12 Genomic alteration detection from long-read, liquid biopsy and single-cell sequencing↑ 2.4×GenomicsRank 13 Accelerated ML methods for fluorescence microscopy and molecular generation↑ 2.0×ML / AI methodsRank 14 Learned scoring ranks fragment-grown molecules by synthesizability and binding affinity↑ 2.0×Structural bio & protein designRank 15 Vision transformers adapted for segmenting micronuclei, plant time-series, and brain MRI↑ 2.0×ML / AI methodsRank 16 Sequence rules and desolvation terms shape condensate phase behavior↑ 1.9×Structural bio & protein designRank 17 Dynamical circuit models link local learning, free-energy spiking and pathway flow↑ 2.0×Systems biologyRank 18 MD simulations map conformational ensembles in ribozyme, actin, and kinesin-5 systems↑ 2.0×Structural bio & protein designRank 19 Open tools quantify motion and aggregation in fluorescence microscopy images↑ 2.0×Tools & infrastructureRank 20 Structured models of pathogen persistence, detection and coevolution in population…↑ 2.0×Systems biology
Show 20 more
Rank 21 Agentic and multi-tool frameworks are orchestrating protein and ligand design workflows↑ 2.0×Structural bio & protein designRank 22 Modeling how proteomic measurements are generated to infer biology from noisy data↑ 1.9×ProteomicsRank 23 LLM agents convert published biomedical data into verifiable structured records↑ 2.0×ML / AI methodsRank 24 Evaluating when learned predictors in molecular and gene discovery can be trusted↑ 2.0×ML / AI methodsRank 25 Scalable genomic criteria refine bacterial taxonomy, plasmids, and membrane proteins↑ 1.9×GenomicsRank 26 Chromosome-level assemblies span nematode, lion, polyploid grass and sorghum genomes↑ 1.9×GenomicsRank 27 Model assumptions determine what evolutionary inference recovers from sequence data↑ 1.8×Systems biologyRank 28 Spiking network models link circuit dynamics to input integration, plasticity, and Φ↑ 2.0×Systems biologyRank 29 Computational inference of spatial cell organization from microscopy and transcriptomics↑ 1.6×Single-cellRank 30 ML-guided peptide design targets fungal pathogens and MRSA persisters↑ 1.6×ML / AI methodsRank 31 Quantum-derived targets drive automated force field and hybrid ML/MM workflows↑ 1.6×Structural bio & protein designRank 32 Quantitative models infer tumor subpopulation evolution, competition, and resistance↑ 1.5×Systems biologyRank 33 Testing when structure-based recovery of biological network dynamics stays reliable↑ 1.6×Systems biologyRank 34 AI-curated and AlphaFold3-based atlases map cardiac and GPCR-G protein interactions↑ 1.5×Structural bio & protein designRank 35 Latent models infer neural dynamics and interactions from large-scale recordings↑ 1.5×Systems biologyRank 36 Segment-Anything models automate morphology measurement in wood and biodiversity images↑ 1.5×Tools & infrastructureRank 37 Computational models resolve immune receptor clonality and antigen-specific T cellsSingle-cellRank 38 Protein language models target function annotation and orphan protein structureGenomicsRank 39 Context-resolved mapping of genetic regulation in pigs, joint tissues and human lncRNAsGenomicsRank 40 Long-read and benchtop platforms extend full-length single-cell RNA profilingSingle-cell