Top 40
View all ›BACKFILLED Code and industry signals are partial for backfilled days.
Rank 1
Genomics
Truth-set benchmarks and T2T references refine variant calling accuracy
Recent work pairs curated truth sets and complete reference genomes with new callers, enabling more rigorous evaluation of variant detection.
2 preprints · 4 published · ↑ 2.3× vs 4-wk avg
Rank 2 ML / AI methods
Deep learning tools for segmenting and classifying biomedical images across modalities
Generalization across cohorts is a central concern, as MICCAI 2024 results show top perivascular segmentation methods degrade on an unseen cohort.
7 preprints · ↑ 2.4× vs 4-wk avg
Rank 3 Single-cell
Histology as an anchor for integrating spatial omics and molecular modalities
Pretrained H&E foundation models are now used as anchors for aligning spatial omics, extending integration beyond purely molecular frameworks.
9 preprints · 9 published · ↑ 2.1× vs 4-wk avg
Rank 4 Systems biology
Optimal control and combination therapy models target bacterial and epidemic persistence
Coupling within-host dynamics, treatment schedules and transmission to population outcomes is a current modeling priority in infection biology.
4 preprints · ↑ 2.3× vs 4-wk avg
Rank 5 Structural bio & protein design
Simulations attribute nucleic acid and nucleosome binding to defined molecular features
Simulations now link binding and partitioning outcomes to specific residues and surface charge across nucleic acid and chromatin systems.
2 preprints · 3 published · ↑ 2.1× vs 4-wk avg
Rank 6 Reference-standard benchmarks expose limits in multi-omics and single-cell methods↑ 2.1×Single-cellRank 7 Software bridges platform-specific omics outputs and downstream statistical analysis↑ 2.2×Tools & infrastructureRank 8 Sequencing tools for RNA modification, integration and somatic variant detection↑ 1.9×GenomicsRank 9 Deep learning models predict drug-target affinity and disease-target associations↑ 1.6×ML / AI methodsRank 10 Molecular networks paired with machine learning to predict cancer phenotypes↑ 1.6×ML / AI methodsRank 11 Individualized brain models infer stable connectivity from fMRI co-activity patterns↑ 1.6×Systems biologyRank 12 Graph neural networks extend to binding affinity, interface, and ADME predictionStructural bio & protein designRank 13 Long-read haplotype-resolved assembly and phasing of heterozygous genomesGenomicsRank 14 Comparative genomes link microbial secretion, metal uptake and extremophile lifestylesGenomicsRank 15 Defense and anti-defense turnover co-evolve across Enterobacteriaceae genomes↑ 1.6×GenomicsRank 16 Spatial indexing and standardized scoring for whole macromolecular assembly comparisonStructural bio & protein designRank 17 Reusable pipelines and models for quality control, imputation, and regulatory genomicsGenomicsRank 18 Adapted protein and RNA models target stability, fitness, and conformational ensemblesStructural bio & protein designRank 19 Machine learning steers antibody design toward affinity and low immunogenicityStructural bio & protein designRank 20 Analysis Pipelines for Impedance and Metabolomic Data With Interpretable MLML / AI methods
Show 20 more
Rank 21 Structure-aware encoders reshape learned representations of DNA, proteins and moleculesML / AI methodsRank 22 ML models predict biosynthetic gene function and enzyme kinetics for strain engineeringML / AI methodsRank 23 Explicit mechanistic models of promoter, metabolic, and stress-response dynamicsSystems biologyRank 24 Structure-derived descriptors probe AlphaFold model quality and domain environmentsStructural bio & protein designRank 25 Single-cell atlases resolve neuronal subtype dynamics and isoform diversityML / AI methodsRank 26 LLM agents orchestrate structure prediction, docking, and biologics or drug designStructural bio & protein designRank 27 Computational models of cancer dynamics, metastasis, and mutational signaturesSystems biologyRank 28 Chromosome-level and pangenome assemblies link genome structure to fatty acid traitsGenomicsRank 29 Polarizable and learned-surface methods extend computational ligand-site predictionStructural bio & protein designRank 30 Spatial methylomes, isoform atlases and imaging cytometry resolve tissue cell statesSingle-cellRank 31 Benchmarks and combined tools for ML prediction in protein function and gene detectionML / AI methodsRank 32 Open-source tools for neuron modeling, brain atlas mapping, and behavior analysisTools & infrastructureRank 33 GPU and k-d tree methods speed up molecular and sequence analysis workloadsTools & infrastructureRank 34 Computational assignment of T-cell antigen recognition and MHC-presented peptidesSingle-cellRank 35 Computational methods sharpen spatial positioning and clone tracing in imaging assaysTranscriptomicsRank 36 R tools pair curated omics and experimental-design data with analysis layersTools & infrastructureRank 37 Inferring cell trajectories and tissue heterogeneity from single-cell transcriptomesSingle-cellRank 38 Mechanical models of cell translocation, epithelial force inference, and cytokinetic flowSystems biologyRank 39 Integrating chromatin contacts, RNA association and binding order in regulatory controlEpigenomicsRank 40 Hi-C-based 3D models resolve ecDNA, nuclear organization and replication timingEpigenomics