A scalable approach to investigating sequence-to-function predictions from personal genomes

Nature Methods 2026 Open access

Read full text
Authors: Anna Spiro, Xinming Tu, Yilun Sheng, Alexander Sasse, Rezwan Hosseini, Maria Chikina, Sara Mostafavi
DOI: 10.1038/s41592-026-03124-8
Cited by: 4
References: 27

Abstract

Sequence-to-function (S2F) models can evaluate arbitrary DNA sequences, yet they struggle to fully capture inter-individual variation in gene expression. We introduce SAGE-net, a scalable framework for training and evaluating S2F models using personal genomes. While personal genome training improves gene expression prediction accuracy for held-out individuals, performance gains arise primarily from identifying predictive variants rather than learning a cis-regulatory grammar that generalizes across loci. Scalable software will be critical to advancing S2F models for personal genomics.

Fields of Study & Subjects

Genomics and Chromatin Dynamics Genomics and Rare Diseases RNA and protein synthesis mechanisms

Publication details

Journal / source
Nature Methods
Publisher
Springer Science and Business Media LLC
Volume
23
Issue
7
Pages
1308–1312
Publication date
8 June 2026
Language
EN