TrendBio

Top 40

View all ›

BACKFILLED Code and industry signals are partial for backfilled days.

Rank 1

ML / AI methods

Pretrained protein models coupled with structural and experimental constraints

Protein language models are now being combined with structure, binding, and experimental data for prediction and design tasks.

6 preprints · 3 published · ↑ 1.7× vs 4-wk avg
Rank 2

Single-cell

Learned representations link omics data to predictive and interpretable readouts

Foundation models, information-bottleneck compression and generative simulators are now being benchmarked and applied across single-cell and spatial omics.

5 preprints · 3 published · ↑ 1.6× vs 4-wk avg
Rank 3

Genomics

Integrative genomics resolves enhancer-promoter wiring and chromatin regulatory states

Orthogonal measurements now resolve regulatory architecture, from single-cell four-omics profiling to GWAS-linked cell states.

5 preprints · 2 published
Rank 4

Structural bio & protein design

MD sampling reveals intermediate-rich landscapes in protein binding and folding

Recent preprints and publications apply MD ensembles and replica-exchange sampling to binding and folding pathways beyond endpoint structures.

3 preprints · 3 published
Rank 5

Genomics

Context-specific error modelling refines somatic and mosaic variant detection

Tumour-informed ctDNA, mosaic STR, and pangenome-based somatic calling studies show gains from modelling information standard pipelines discard.

5 preprints · 1 published
Rank 6 New software tools automate metagenomic and mobile-element sequence analysis stepsTools & infrastructureRank 7 New tools for gene IDs, proteomics, sequencing QC, and microbial protein annotationTools & infrastructureRank 8 Computational tools mine microbial sequence, metagenomic, and metabolic dataTools & infrastructureRank 9 Re-analysis recovers TF signatures from pooled single-cell screens lacking controlsSingle-cellRank 10 Generative models reconstruct unobserved single-cell state transitions across modalitiesSingle-cellRank 11 Validating dated phylogenies, imputation, and random forest genomic predictionGenomicsRank 12 Automated k-mer and learned-model tools streamline sequencing and phylogenomic analysisGenomicsRank 13 Neural dynamics and connectivity probed by bifurcation, recurrence and dimension methodsSystems biologyRank 14 Learned and knowledge-based potentials score ligand, glycan and metal bindingStructural bio & protein designRank 15 Protein language models predict bacterial defense systems and virus-host interactionsML / AI methodsRank 16 Pretrained models learn transferable representations of counts, calcium traces, and cellsML / AI methodsRank 17 Phase and frequency of EEG oscillations characterized as dynamical quantitiesSystems biologyRank 18 Spatial and localization data expose RNA-protein and metabolic heterogeneity in tumorsMetabolomicsRank 19 Graph-based representation learning for biological association and single-cell dataML / AI methodsRank 20 Chromosome-scale assemblies drive demographic, comparative and sex chromosome inferenceGenomics