Uncertainty-aware genomic deep learning with knowledge distillation

Zhou, Jessica, Rizzo, Kaeli, Christensen, Trevor, Tang, Ziqi, Koo, Peter K (2026) Uncertainty-aware genomic deep learning with knowledge distillation. npj Artificial Intelligence, 2 (1). p. 3. ISSN 3005-1460 (Public Dataset)

[thumbnail of 10.1038.s44387-025-00053-3.pdf]

PDF
10.1038.s44387-025-00053-3.pdf - Published Version
Available under License Creative Commons Attribution Non-commercial No Derivatives.
Download (21MB)

URL: https://www.ncbi.nlm.nih.gov/pubmed/41523993

DOI: 10.1038/s44387-025-00053-3

Abstract

Deep neural networks (DNNs) have advanced predictive modeling for regulatory genomics, but challenges remain in ensuring the reliability of their predictions and understanding the key factors behind their decision making. Here we introduce DEGU (Distilling Ensembles for Genomic Uncertainty-aware models), a method that integrates ensemble learning and knowledge distillation to improve the robustness and explainability of DNN predictions. DEGU distills the predictions of an ensemble of DNNs into a single model, capturing both the average of the ensemble's predictions and the variability across them, with the latter representing epistemic (or model-based) uncertainty. DEGU also includes an optional auxiliary task to estimate aleatoric, or data-based, uncertainty by modeling variability across experimental replicates. By applying DEGU across various functional genomic prediction tasks, we demonstrate that DEGU-trained models inherit the performance benefits of ensembles in a single model, with improved generalization to out-of-distribution sequences and more consistent explanations of cis-regulatory mechanisms through attribution analysis. Moreover, DEGU-trained models provide calibrated uncertainty estimates, with conformal prediction offering coverage guarantees under minimal assumptions. Overall, DEGU paves the way for robust and trustworthy applications of deep learning in genomics research.

Item Type:	Paper
Subjects:	bioinformatics bioinformatics > quantitative biology
CSHL Authors:	Tang, Ziqi Koo, Peter K Zhou, Jessica Rizzo, Kaeli Christensen, Trevor
Communities:	CSHL labs > Koo Lab
SWORD Depositor:	CSHL Elements
Depositing User:	CSHL Elements
Date:	2026
Date Deposited:	13 Jan 2026 14:29
Last Modified:	13 Jan 2026 14:29
PMCID:	PMC12779563
Related URLs:	Publisher
Dataset ID:	https://doi.org/10.5281/zenodo.14145284
URI:	https://repository.cshl.edu/id/eprint/42066

Actions (login required)

Administrator's edit/view item