TrendBio

Top 40

View all ›

BACKFILLED Code and industry signals are partial for backfilled days.

Rank 1

Transcriptomics

Graph-based learning extends spatial transcriptomics to imputation, denoising and 3D

Paired histology and spatial expression data now support graph and implicit models for imputation, denoising, and 3D reconstruction.

13 preprints · 8 published · ↑ 3.0× vs 4-wk avg
Rank 2

ML / AI methods

Omics machine learning benchmarks test generalization across populations and modalities

Recent benchmarks shift omics model evaluation toward subgroup disparities, cross-modality transfer, and large-scale pipeline comparisons.

6 preprints · 9 published · ↑ 2.3× vs 4-wk avg
Rank 3

ML / AI methods

Sequence-based deep models predict gene regulation and chromatin features from DNA

Deep sequence models are now being tested for regulatory predictability, methylation context and chromatin looping, not just expression prediction.

7 preprints · 5 published · ↑ 1.9× vs 4-wk avg
Rank 4

Tools & infrastructure

Open-source software for methylation, cancer genomics, and regulatory analysis

Recent June 2026 releases pair reproducible Snakemake workflows with LLM-assisted metadata extraction and new imputation and mapping methods.

1 preprint · 10 published · ↑ 1.8× vs 4-wk avg
Rank 5

Structural bio & protein design

Benchmarks expose limits of learned models for disordered proteins and complexes

Recent work benchmarks structure predictors and foundation models against experimental and simulation data.

8 preprints · 2 published · ↑ 1.6× vs 4-wk avg