TrendBio

Top 40

View all ›

BACKFILLED Code and industry signals are partial for backfilled days.

Rank 1

Transcriptomics

Computational methods place gene expression and cell arrangement within tissue geometry

Spatial omics methods now address 3D reconstruction, differential co-localisation, and subcellular mRNA localization in tissue context.

7 preprints · 7 published · ↑ 4.5× vs 4-wk avg
Rank 2

Single-cell

Atlases attribute genetic risk and developmental state to individual cell types

Harmonized human single-cell atlases now span millions of cells, enabling uniform cross-dataset comparison of genetic and developmental effects at cell-type…

5 preprints · 2 published · ↑ 3.3× vs 4-wk avg
Rank 3

Structural bio & protein design

RL-steered generative models expand controllable molecule and antibody design

Recent preprints and papers couple generative models with reinforcement learning and Bayesian optimization for explicit design goals.

4 preprints · 5 published · ↑ 3.0× vs 4-wk avg
Rank 4

Systems biology

Perturbation screens coupled to genome, morphology and transcriptome readouts

Screens now pair large perturbation sets with multimodal or single-cell readouts across neural, cardiac, and melanoma models.

3 preprints · 3 published · ↑ 3.0× vs 4-wk avg
Rank 5

ML / AI methods

Metabolic models and protein language models predict microbial and phage traits

Genome-scale metabolic models and protein language models now target interaction specificity and intervention targets, not only sequence annotation.

4 preprints · 2 published · ↑ 3.0× vs 4-wk avg
Rank 6 Computational predictors and biophysical models quantify how variants disrupt function↑ 3.0×GenomicsRank 7 Machine learning links tissue and molecular profiles to clinical outcome endpoints↑ 2.6×ML / AI methodsRank 8 ML models predict antibody and antigen presentation from sequence and structure↑ 2.5×ML / AI methodsRank 9 Ortholog and comparative genomics tools applied across bacterial and model systems↑ 2.1×Tools & infrastructureRank 10 Docking and fingerprint methods prioritize ligands, binding modes, and protein targets↑ 2.3×Structural bio & protein designRank 11 Learned models map DNA sequence and omic modalities onto chromatin and contact profiles↑ 2.3×EpigenomicsRank 12 Generative models sample protein and antibody conformations via learned latent spaces↑ 2.2×Structural bio & protein designRank 13 Adhesion and polarity noise constrain collective migration in mechanical cell models↑ 2.1×Systems biologyRank 14 Explicit latent structure for single-cell perturbation response and multi-omic modeling↑ 2.0×Single-cellRank 15 Graph and attention models predict drug-target and drug-disease associations↑ 1.9×ML / AI methodsRank 16 Models link non-genetic B cell state dynamics to antibody affinity maturation↑ 2.0×ML / AI methodsRank 17 Software layers scoring and visualization onto peptide and cross-link resources↑ 2.0×ProteomicsRank 18 Input depth and scoring criteria govern accuracy in protein complex and MD analyses↑ 1.8×Structural bio & protein designRank 19 Theoretical models frame cellular organization as statistically characterized dynamics↑ 2.0×Systems biologyRank 20 LLMs reasoning over metabolic networks and public metabolomics data↑ 1.6×Metabolomics