Benchmarking DNA Foundation Models for Genomic and Genetic Tasks

Authors

Haonan Feng
Lang Wu
Bingxin Zhao
Chad Huff
Jianjun Zhang
Jia Wu
Lifeng Lin
Peng Wei
Chong Wu

Language

English

Publication Date

11-28-2025

Journal

Nature Communications

DOI

10.1038/s41467-025-65823-8

PMID

41315262

PMCID

PMC12663285

PubMedCentral® Posted Date

11-28-2025

PubMedCentral® Full Text Version

Post-print

Abstract

The rapid evolution of DNA foundation models promises to revolutionize genomics, yet comprehensive evaluations are lacking. Here, we present a comprehensive, unbiased benchmark of five models (DNABERT-2, Nucleotide Transformer V2, HyenaDNA, Caduceus-Ph, and GROVER) across diverse genomic and genetic tasks including sequence classification, gene expression prediction, variant effect quantification, and topologically associating domain (TAD) region recognition, using zero-shot embeddings. Our analysis reveals that mean token embedding consistently and significantly improves sequence classification performance, outperforming other pooling strategies. Model performance varies among tasks and datasets; while general purpose DNA foundation models showed competitive performance in pathogenic variant identification, they were less effective in predicting gene expression and identifying putative causal QTLs compared to specialized models. Our findings offer a framework for model selection, highlighting the impact of architecture, pre-training data, and embedding strategies on performance in genomic and genetic tasks.

Keywords

Genomics, Benchmarking, Humans, Models, Genetic, DNA, Quantitative Trait Loci, Sequence Analysis, DNA, Genomics, Quantitative trait, Genome, Computational models, Machine learning

Published Open-Access

yes

Recommended Citation

Feng, Haonan; Wu, Lang; Zhao, Bingxin; et al., "Benchmarking DNA Foundation Models for Genomic and Genetic Tasks" (2025). Faculty, Staff and Student Publications. 5884.
https://digitalcommons.library.tmc.edu/uthgsbs_docs/5884