GEO series

Evolutionary transfer learning enables organism-wide inference of mammalian enhancer landscapes

GSE325776 Mus musculus Genome binding/occupancy profiling by high throughput sequencing 36 samples Submitted 2026/04/09 Platform GPL24247

Summary

Understanding and modeling how a single human genome concurrently encodes gene regulatory programs for thousands of cell types remains a central challenge in genomics and machine learning. Most human cell types emerge during embryonic, fetal, and pediatric development which are inaccessible to comprehensive molecular profiling. To circumvent this, we hypothesized that the mismatch in evolutionary rates between cis-acting enhancers that modulate gene expression (fast) and the trans-acting regulatory factors that specify cell types (slow) creates an opportunity for 'evolutionary transfer learning'. Specifically, models trained to predict cell type-specific enhancers in one species should generalize to the orthologous cell types and enhancers of related species. To test this, we generated a single-cell atlas of chromatin accessibility spanning mouse embryonic day 10 (E10) to birth (P0). Using combinatorial indexing, we profiled 3.9 million nuclei from 36 staged embryos, resolving genome-wide accessibility in 36 cell classes and 140 cell types. We then trained a series of multi-output deep learning models (CREsted), each addressing limitations of the preceding approach, towards the goal of genome-wide prediction of distal enhancers across major developmental lineages. An 'evolution-naive' model achieved strong performance on heldout peaks, but exhibited two failure modes during genome-wide inference: overprediction at tandem repeats and conflation of promoter and distal enhancer grammars. An 'evolution-aware' model resolved these by regrouping accessible regions based on their retention and functional coherence across mammalian evolution, but failed to generalize across species. Finally, an 'evolution-augmented' model, STEAM (Synteny-aware Transfer learning for Enhancer Activity Modeling), incorporated enhancer orthologs from 241 mammalian genomes (Zoonomia) in a synteny-supervised manner. This increased the effective data scale by as much as 195-fold, markedly improving generalization across mammals despite greater label noise. We applied STEAM to the genome-wide inference of cell class-specific distal developmental enhancers in humans, mice (HumMus) and 239 additional mammals (BabaGanoush), i.e. 32 × 241 = 7,712 genome-wide distal enhancer prediction tracks. Together, our results unify advances in single-cell profiling, deep learning, and comparative genomics into a framework for the evolutionary transfer learning of noncoding regulatory grammars. More broadly, our work supports the view that model organisms and evolutionarily diverse genomes are indispensable resources for accelerating and enhancing the AI-enabled exploration of human biology.

Published in

Qiu C, Daza RM, Welsh IC et al. · bioRxiv : the preprint server for biology 2026 · PMID 41993329 · doi:10.64898/2026.04.07.717039

This dataset

Open GSE325776 in finder → Find more mouse ChIP / ATAC / CUT&Tag datasets →

Download

Direct links to NCBI, no account and no request form: the whole study as GSE325776_RAW.tar, processed values as the series matrix, the supplementary file directory, and per-sample supplementary files for any of the 36 samples. Raw sequencing reads are also available from ENA.

Whole study · RAW.tar ↓ Series matrix ↓ All supplementary files ↗ Raw reads on ENA ↗

Also filed as BioProject PRJNA1441597. Searching any of these in the dataset finder brings you back here.

NCBI GEO page ↗ Paper (PMID 41993329) ↗

Study design

36 conditions, each sampled once — no replicated groups

Read from 36 sample titles: 36 distinct titles with little repetition. Check it against the sample list below before relying on it.

Similar datasets

Search all mouse ChIP / ATAC / CUT&Tag datasets in GEO →

Share this dataset

Metadata from NCBI GEO, cached and refreshed periodically — the NCBI page above is authoritative. Downloads link straight to NCBI/ENA; nothing is proxied through BioTransfer.