DREAM-DNA: Controlled Design via Reasoning and Matched-flows for DNA

Weimin Wu, Jiunhau Chen, Xuefeng Song, Jerry Yao-Chieh Hu, Han Liu

Track: Full / long paper (5-8 pages)

Keywords: Genome, Flow-matching, Reinforcement Learning

TL;DR: A reinforcement learning–based generative framework for DNA sequence design

Abstract:

We introduce DREAM-DNA (controlled Design via REasoning And Matched-flows for DNA), a reinforcement learning–based generative framework for DNA sequence design.
Traditional models in this domain use diffusion architecture and rely on stochastic, single-pass generation.
These limit their ability to correct early structural errors.
DREAM-DNA overcomes these limitations by replacing stochasticity with Discrete Flow Matching.
This establishes a deterministic ''straight-line'' mapping from noise to data for superior trajectory control.
Our framework further introduces an Iterative Rollout mechanism.
Specifically, we treat generation as a multi-round refinement process.
By iteratively masking and resampling subsets of nucleotides in the DNA sequence, the model ``critiques'' and revises its intermediate outputs based on the reward feedback.
Experiments on human enhancer design show that DREAM-DNA outperforms state-of-the-art baselines.
It achieves a 13% boost in enhancer activity and an 8% increase in open chromatin match levels.

Anonymization: This submission has been anonymized for double-blind review via the removal of identifying information such as names, affiliations, and identifying URLs.

Submission Number: 14