DNA-BFN: Bayesian flow networks for generative modeling of regulatory DNA sequences
Abstract
Motivation: Designing regulatory DNA sequences is challenging due to their discrete nucleotide alphabet and complex higher-order dependencies underlying regulatory function. Many existing generative models, particularly diffusion-based approaches, learn by injecting stochastic noise into sequences and training denoising processes, which can distort discrete sequence structure and weaken biological constraints.
Results: Here, we introduce DNA Bayesian Flow Networks (DNA-BFN), a generative framework that models regulatory DNA sequences through iterative Bayesian updates of per-position nucleotide distributions. By directly modeling per-position categorical beliefs, DNA-BFN realizes a diffusion-like generative process without explicitly perturbing discrete nucleotide sequences, enabling fully differentiable training while preserving discrete sequence structure. Across eight regulatory sequence datasets spanning diverse species, scales, and sequence lengths, DNA-BFN consistently outperforms state-of-the-art baselines, faithfully preserves sequence feature statistics and regulatory patterns, enables few-shot sequence design, and achieves faster sampling efficiency. These results show that framing sequence generation as progressive uncertainty reduction in probability space provides an effective and generalizable approach for nucleotide sequence design.
Availability and implementation: The software and its source code are publicly available at https://github.com/WangLabTHU/DNA-BFN. The dataset, source code and model weights are available in the Zenodo repository: https://doi.org/10.5281/zenodo.20177972.
Related
- CardamomOT: A mechanistic optimal transport-based framework for gene regulatory network inference, trajectory reconstruction and generative modeling
- RNAGEN: A Generative Adversarial Network‐Based Model to Generate Synthetic RNA Sequences to Target Proteins
- An integrative multi-omics modeling framework for chemotherapy response prediction through microRNA–target regulatory signatures
- Spatiotemporal Multi-Omics and Compartment-Resolved Metabolomic Analyses Reveal the Mechanisms and Core Regulatory Networks of Cuticular Wax Formation in Blueberry Fruits
- Dynamic three-dimensional chromatin remodeling in lung disease: mechanistic principles, pathogenic regulatory networks, and translational challenges
BibTeX
@article{yu2026dnabfn,
title = {{DNA-BFN: Bayesian flow networks for generative modeling of regulatory DNA sequences}},
author = {Han Yu and Qixiu Du and Lei Wei and Xiaowo Wang},
journal = {Bioinformatics},
year = {2026},
month = oct,
doi = {10.1093/bioinformatics/btag785},
url = {https://doi.org/10.1093/bioinformatics/btag785},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - DNA-BFN: Bayesian flow networks for generative modeling of regulatory DNA sequences
AU - Han Yu
AU - Qixiu Du
AU - Lei Wei
AU - Xiaowo Wang
PY - 2026
DA - 2026/10/19
JO - Bioinformatics
DO - 10.1093/bioinformatics/btag785
UR - https://doi.org/10.1093/bioinformatics/btag785
AB - Motivation: Designing regulatory DNA sequences is challenging due to their discrete nucleotide alphabet and complex higher-order dependencies underlying regulatory function. Many existing generative models, particularly diffusion-based approaches, learn by injecting stochastic noise into sequences and training denoising processes, which can distort discrete sequence structure and weaken biological constraints. Results: Here, we introduce DNA Bayesian Flow Networks (DNA-BFN), a generative framework that models regulatory DNA sequences through iterative Bayesian updates of per-position nucleotide distributions. By directly modeling per-position categorical beliefs, DNA-BFN realizes a diffusion-like generative process without explicitly perturbing discrete nucleotide sequences, enabling fully differentiable training while preserving discrete sequence structure. Across eight regulatory sequence datasets spanning diverse species, scales, and sequence lengths, DNA-BFN consistently outperforms state-of-the-art baselines, faithfully preserves sequence feature statistics and regulatory patterns, enables few-shot sequence design, and achieves faster sampling efficiency. These results show that framing sequence generation as progressive uncertainty reduction in probability space provides an effective and generalizable approach for nucleotide sequence design. Availability and implementation: The software and its source code are publicly available at https://github.com/WangLabTHU/DNA-BFN. The dataset, source code and model weights are available in the Zenodo repository: https://doi.org/10.5281/zenodo.20177972.
SN - 1367-4803
SN - 1367-4811
ER -