R-loop Prediction Reveals Generalization Limits of DNA Foundation Models Beyond Regulatory Genomics
Preprint
Abstract
DNA foundation models are increasingly proposed as general-purpose representations for genomic prediction and design, yet their evaluation remains largely centered on conventional regulatory tasks. This leaves a critical question unresolved: do DNA foundation models generalize to sequence biology beyond conventional gene regulation? To answer this question, we introduce RloopBench , a systematic benchmark for R-loop-forming sequence prediction as a biophysically distinct, genome-stability-associated task. We compare rule-based methods, task-specific models, classical sequence encodings, and foundation model representations across in-distribution, cross-platform, consensus-level, and cross-species evaluations. Foundation models achieve strong performance when positive and negative sequences are compositionally separable, but this advantage does not consistently transfer to cross-platform and cross-species settings, where they are often comparable to classical k -mer representations. Unexpectedly, a one-hot classifier baseline shows the strongest overall sensitivity to R-loop-forming sequences, exceeding more complex models across several generalization tests. Rule-based and task-specific models also exhibit limited transfer outside their original training regimes. Performance is further shaped by sequence properties, negative-control design, experimental platform, and species-specific genomic context. Together, RloopBench establishes genome-stability-associated sequence prediction as a complementary direction for DNA foundation model development and evaluation, while underscoring that simple sequence encodings remain necessary baselines for assessing model generalization beyond conventional regulatory tasks.Keywords
Funding Information
- NC State Start-up Fund
- National Science Foundation (MCB-2540505)
- NC State Comparative Medicine Institute Seed Fund
- NC State Genetics and Genomics Academy Seed Fund
This publication has 57 references indexed in Scilit:
- Base-pair resolution reveals clustered R-loops and DNA damage-susceptible R-loopsMolecular Cell, 2025
- Deep learning-enhanced R-loop prediction provides mechanistic implications for repeat expansion diseasesiScience, 2024
- ULI-ssDRIP-seq revealed R-loop dynamics during vertebrate early embryogenesisCell Insight, 2024
- A survey of k-mer methods and applications in bioinformaticsComputational and Structural Biotechnology Journal, 2024
- Approaches for Mapping and Analysis of R‐loopsCurrent Protocols, 2024
- G-quadruplexes associated with R-loops promote CTCF bindingMolecular Cell, 2023
- Defining R-loop classes and their contributions to genome instabilityDNA Repair, 2021
- The Augmented R-Loop Is a Unifying Mechanism for Myelodysplastic Syndromes Induced by High-Risk Splicing Factor MutationsMolecular Cell, 2018
- R-ChIP Using Inactive RNase H Reveals Dynamic Coupling of R-loops with Transcriptional Pausing at Gene PromotersMolecular Cell, 2017
- Prevalent, Dynamic, and Conserved R-Loop Structures Associate with Specific Epigenomic Signatures in MammalsMolecular Cell, 2016