Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Research Highlight
  • Published:

Transcriptomics

A chromatin-based model for deciphering gene interactions

Nature Biotechnology volume 43, page 176 (2025)Cite this article

This article has been updated

Foundation models have leveraged diverse gene expression profiles for tasks such as cell type annotation or prediction of perturbation response. However, no model yet exists to demonstrate how the chromatin landscape regulates transcription. Writing in Nature, Fu et al. introduce the general expression transformer (GET), a foundation model trained on chromatin accessibility data from diverse human cell types, and use it to predict gene activity within specific cells.

The GET model was pretrained using single-cell chromatin accessibility data from 213 human cell types and subsequently fine-tuned on paired chromatin and RNA sequencing data, allowing it to learn how regions and features interact with each other across diverse cell types. In fetal erythroblasts, GET identified long-range cis-regulatory elements and their associated transcription factors that were missed by previous models. The researchers also constructed the publicly available GET Catalog, which combines GET-predicted regulatory sequences and protein structure information from AlphaFold2 to predict direct or indirect transcription factor interactions. They use the catalog to identify how the lymphocyte-specific PAX5 and nuclear receptor interaction might be enhanced by a specific PAX5 germline mutation associated with increased risk of B cell acute lymphoblastic leukemia.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on SpringerLink
  • Instant access to the full article PDF.

USD 39.95

Prices may be subject to local taxes which are calculated during checkout

Change history

  • 06 May 2025

    In the version of the article initially published, "Fu et al." incorrectly appeared as “Lu et al.” and has now been amended in the HTML and PDF versions of the article.

Author information

Authors and Affiliations

  1. Nature Biotechnology https://www.nature.com/nbt/

    Rong Li

Authors
  1. Rong Li
    View author publications

    Search author on:PubMed Google Scholar

Corresponding author

Correspondence to Rong Li.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Li, R. A chromatin-based model for deciphering gene interactions. Nat Biotechnol 43, 176 (2025). https://doi.org/10.1038/s41587-025-02573-4

Download citation

  • Published:

  • Version of record:

  • Issue date:

  • DOI: https://doi.org/10.1038/s41587-025-02573-4

Search

Advanced search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing