TrendBio

Top 40

View all ›

BACKFILLED Code and industry signals are partial for backfilled days.

Rank 1

Structural bio & protein design

Generative models steered toward objectives for molecules, proteins and antibodies

Multi-objective search and reinforcement learning are now routinely steering generative models toward explicit design objectives.

9 preprints · 5 published · ↑ 3.8× vs 4-wk avg
Rank 2

Transcriptomics

Spatial omics methods map cell neighbourhoods, mRNA localization and 3D tissue structure

Spatial profiling is moving from single sections toward volumetric, subcellular and multi-sample analyses of intact tissue.

3 preprints · 3 published · ↑ 3.0× vs 4-wk avg
Rank 3

Epigenomics

Learned models map sequence to chromatin contacts and multi-omic single-cell states

Recent preprints and publications apply learned models across cross-species ChIP-seq, Hi-C, scATAC-seq and multi-omic data, with explicit benchmarking.

4 preprints · 7 published · ↑ 2.9× vs 4-wk avg
Rank 4

Genomics

Variant effects quantified across protein stability, DNA binding, and gene dosage

Recent work benchmarks indel predictors against clinical evidence levels while quantifying conformational, dosage, and developmental consequences of variants.

2 preprints · 3 published · ↑ 2.7× vs 4-wk avg
Rank 5

ML / AI methods

ML maps tissue and molecular profiles to tumor outcomes and clonal evolution

Deep learning now links H&E, spatial proteomics, and single-cell data to clinical outcomes and tumor evolution endpoints.

5 preprints · 6 published · ↑ 2.5× vs 4-wk avg
Rank 6 ML models predict immune receptor and antigen presentation from sequence data↑ 2.5×ML / AI methodsRank 7 Metabolic models and protein language models predict pathogen and phage traits↑ 2.3×ML / AI methodsRank 8 Models trained on large binding datasets guide ligand and binder discovery↑ 2.4×Structural bio & protein designRank 9 Automated and single-cell platforms map genome architecture and perturbation phenotypes↑ 2.3×EpigenomicsRank 10 Pairwise and hierarchical structure in omics data made interpretable computationally↑ 2.3×ML / AI methodsRank 11 Computational models infer metabolic kinetics, microbial competition and proteome state↑ 2.3×Systems biologyRank 12 Packaged tools make proteomics, ortholog and metagenomic resources directly usable↑ 2.3×Tools & infrastructureRank 13 Quantitative dynamics model how intracellular structures are built and maintained↑ 2.3×Systems biologyRank 14 Pretrained generative models extend to fMRI forecasting, diagnosis, and imagery↑ 2.1×ML / AI methodsRank 15 Generative models map mass spectra directly to candidate molecular structures↑ 2.0×MetabolomicsRank 16 Structured latent spaces for single-cell perturbation and representation learning↑ 2.0×Single-cellRank 17 Graph storage and streaming parsers reshape population-scale variant computation↑ 2.0×Tools & infrastructureRank 18 Mathematical models link noise, feedback, and phase behavior to cell fate and dormancy↑ 1.9×Systems biologyRank 19 Structural modeling of disorder and interfaces predicts protein-protein interactions↑ 1.7×Structural bio & protein designRank 20 Benchmarks probe uncertainty in RNA-seq and nanopore direct RNA analysis tools↑ 1.7×Transcriptomics