DART-Eval: A Comprehensive DNA Language Model Evaluation Benchmark on Regulatory DNA

TL;DR

DART-Eval introduces a comprehensive benchmark suite for evaluating DNA language models on regulatory DNA tasks, revealing current models' inconsistent performance and high resource demands, guiding future improvements.

Contribution

This work provides the first dedicated benchmark for regulatory DNA tasks, enabling systematic assessment of DNALMs across multiple scenarios and tasks.

Findings

Current DNALMs show inconsistent performance.

DNALMs do not outperform baseline models on most tasks.

DNALMs require significantly more computational resources.

Abstract

Recent advances in self-supervised models for natural language, vision, and protein sequences have inspired the development of large genomic DNA language models (DNALMs). These models aim to learn generalizable representations of diverse DNA elements, potentially enabling various genomic prediction, interpretation and design tasks. Despite their potential, existing benchmarks do not adequately assess the capabilities of DNALMs on key downstream applications involving an important class of non-coding DNA elements critical for regulating gene activity. In this study, we introduce DART-Eval, a suite of representative benchmarks specifically focused on regulatory DNA to evaluate model performance across zero-shot, probed, and fine-tuned scenarios against contemporary ab initio models as baselines. Our benchmarks target biologically meaningful downstream tasks such as functional sequence feature discovery, predicting cell-type specific regulatory activity, and counterfactual prediction of the impacts of genetic variants. We find that current DNALMs exhibit inconsistent performance and do not offer compelling gains over alternative baseline models for most tasks, while requiring significantly more computational resources. We discuss potentially promising modeling, data curation, and evaluation strategies for the next generation of DNALMs. Our code is available at https://github.com/kundajelab/DART-Eval.