TrendBio

Top 40

View all ›

BACKFILLED Code and industry signals are partial for backfilled days.

Rank 1

Structural bio & protein design

Deep learning predicts protein binding, disorder, phase separation and ligand poses

Preprints and papers now pair protein language models and graph networks with curated structural data for binding and disorder prediction.

4 preprints · 2 published · ↑ 1.5× vs 4-wk avg
Rank 2

Tools & infrastructure

Protein language models outperform BLAST for distant enzyme function prediction

Recent preprints and papers benchmark protein language models against BLASTp and structure-based search for enzyme and microbial protein annotation.

3 preprints · 2 published
Rank 3

Genomics

Subcellular localization and cell-type matching refine omics data interpretation

Recent preprints and a published method address interpretive confounds in transcriptomic, proteomic, and metabolomic profiling.

4 preprints · 1 published
Rank 4

ML / AI methods

ML models decode regulatory wiring, aging trajectories and single-cell omics structure

Multimodal omics and temporal foundation models now infer regulatory drivers of cell state, not only describe them.

2 preprints · 3 published
Rank 5

Tools & infrastructure

New tools and methods for microbiome profiling and mobile element analysis from reads

Recent preprints and a published method target depth-dependent microbiome profiling and open visualization of genome and transductomics data.

3 preprints · 1 published
Rank 6 Simulations map protein-peptide binding pathways and peptide folding landscapesStructural bio & protein designRank 7 Pangenome and imputation resources refine germline variant detection across populationsGenomicsRank 8 Machine learning methods infer selection and sample origin from genetic dataGenomicsRank 9 Phage infection reshapes host transcription and metabolism across datasets and modelsTools & infrastructureRank 10 Mammal and avian genomes test neutral theory, fluctuating selection, and diversity lossGenomicsRank 11 Pangenome references and STR-aware models refine mosaic and somatic variant callingGenomicsRank 12 ML models for MHC-I immunopeptidomics, TCR:pMHC binding, and peptide designML / AI methods