TrendBio

Top 40

View all ›

BACKFILLED Code and industry signals are partial for backfilled days.

Rank 1

Single-cell

Learned representations are resolving spatial and single-cell omics at multiple scales

Spatial and single-cell datasets are growing in scale and modality, prompting foundation models, QC frameworks and deconvolution tools.

9 preprints · 5 published · ↑ 2.8× vs 4-wk avg
Rank 2

Structural bio & protein design

Pretrained structure models repurposed for generative protein and peptide design

Diffusion, score-distillation, and RL-based design methods now target de novo binders and pockets with explicit physical or preference constraints.

8 preprints · 2 published · ↑ 1.8× vs 4-wk avg
Rank 3

Structural bio & protein design

RNA structure methods pair folding models with geometric, template, and raw-signal data

Blind benchmarking and new geometric and nanopore datasets now test RNA structure methods beyond sequence alone.

6 preprints · ↑ 1.8× vs 4-wk avg
Rank 4

ML / AI methods

Supervised and transfer-learning models classify cancer subtypes from multi-omics data

Transfer learning, meta-learning, and ensemble stacking are being applied to heterogeneous omics inputs for subtype and outcome prediction.

3 preprints · 4 published · ↑ 1.7× vs 4-wk avg
Rank 5

Structural bio & protein design

MD and ML approaches resolve conformational states, solvation and selectivity

Recent MD studies separate residue-specific contacts from net charge and resolve solvation and selectivity determinants at atomistic detail.

5 preprints · 1 published · ↑ 1.5× vs 4-wk avg
Rank 6 Comparative genomics traces gene family gain and loss across mammals and plants↑ 1.5×GenomicsRank 7 Large-scale genome mining links sequence content to defined microbial systems and traitsGenomicsRank 8 Graph networks predict drug-target affinity, synergy, and gene-pair interactionsML / AI methodsRank 9 Deep learning recovers latent temporal and regulatory structure from static omics profilesSingle-cellRank 10 AI models predict drug synergy, epitope immunogenicity, and viral infectivityML / AI methodsRank 11 Attention and pretrained genomic models target bacterial resistance and interactionsML / AI methodsRank 12 Multilocus models refine genetic architecture from summary and sequencing dataGenomicsRank 13 Ensemble and network-aware models extend protein prediction across microbial dataTools & infrastructureRank 14 Data-driven tools decode mutations, mutational signatures, and T-ALL leukemia subtypesGenomicsRank 15 Confinement drives chiral rotation; epigenetic feedback reshapes cell-state landscapesSingle-cellRank 16 Methods for estimating complex trait architecture from GWAS summary and biomarker dataGenomicsRank 17 DNA sequence encodes enhancers, replication origins and nucleosome occupancyEpigenomicsRank 18 Primer and amplicon workflows refine targeted sequencing enrichment and read resolutionGenomicsRank 19 Protein AI models expand conformer sampling and in silico evolution; developability lagsStructural bio & protein designRank 20 Conditional generative models target pathogen-specific antimicrobial peptide designML / AI methods
Show 10 moreShow less
Rank 21 ML priors now guide cryo-EM reconstruction and electron microscopy acquisitionStructural bio & protein designRank 22 Computable models of tissue transport, signaling, and mechanics in multicellular systemsSystems biologyRank 23 Latent dynamical models of neural activity are extended and judged beyond predictionSystems biologyRank 24 Cell-type-resolved atlases map expression and splicing variation across mouse and humanSingle-cellRank 25 Population variants and sequence diversity shape CRISPR guide and construct designML / AI methodsRank 26 Benchmarking and deep learning for microbiome network, phage-host, and disease inferenceML / AI methodsRank 27 Agents chain target discovery, pocket design, and PK simulation for therapeutic designML / AI methodsRank 28 Reproducible software infrastructure packaging curated biological datasets for analysisTools & infrastructureRank 29 Interpretable factorization and feature curation refine single-cell representationsSingle-cellRank 30 Imposing explicit causal and design structure on high-dimensional molecular dataSystems biology

Sponsors