Ctrl-DNA: Controllable Cell-Type-Specific Regulatory DNA Design via Constrained RL

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Xingyu, Ma, Shihao, Lin, Runsheng, Lin, Jiecong, Wang, Bo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913860015882240
author Chen, Xingyu
Ma, Shihao
Lin, Runsheng
Lin, Jiecong
Wang, Bo
author_facet Chen, Xingyu
Ma, Shihao
Lin, Runsheng
Lin, Jiecong
Wang, Bo
contents Designing regulatory DNA sequences that achieve precise cell-type-specific gene expression is crucial for advancements in synthetic biology, gene therapy and precision medicine. Although transformer-based language models (LMs) can effectively capture patterns in regulatory DNA, their generative approaches often struggle to produce novel sequences with reliable cell-specific activity. Here, we introduce Ctrl-DNA, a novel constrained reinforcement learning (RL) framework tailored for designing regulatory DNA sequences with controllable cell-type specificity. By formulating regulatory sequence design as a biologically informed constrained optimization problem, we apply RL to autoregressive genomic LMs, enabling the models to iteratively refine sequences that maximize regulatory activity in targeted cell types while constraining off-target effects. Our evaluation on human promoters and enhancers demonstrates that Ctrl-DNA consistently outperforms existing generative and RL-based approaches, generating high-fitness regulatory sequences and achieving state-of-the-art cell-type specificity. Moreover, Ctrl-DNA-generated sequences capture key cell-type-specific transcription factor binding sites (TFBS), short DNA motifs recognized by regulatory proteins that control gene expression, demonstrating the biological plausibility of the generated sequences.
format Preprint
id arxiv_https___arxiv_org_abs_2505_20578
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Ctrl-DNA: Controllable Cell-Type-Specific Regulatory DNA Design via Constrained RL
Chen, Xingyu
Ma, Shihao
Lin, Runsheng
Lin, Jiecong
Wang, Bo
Machine Learning
Artificial Intelligence
Genomics
Designing regulatory DNA sequences that achieve precise cell-type-specific gene expression is crucial for advancements in synthetic biology, gene therapy and precision medicine. Although transformer-based language models (LMs) can effectively capture patterns in regulatory DNA, their generative approaches often struggle to produce novel sequences with reliable cell-specific activity. Here, we introduce Ctrl-DNA, a novel constrained reinforcement learning (RL) framework tailored for designing regulatory DNA sequences with controllable cell-type specificity. By formulating regulatory sequence design as a biologically informed constrained optimization problem, we apply RL to autoregressive genomic LMs, enabling the models to iteratively refine sequences that maximize regulatory activity in targeted cell types while constraining off-target effects. Our evaluation on human promoters and enhancers demonstrates that Ctrl-DNA consistently outperforms existing generative and RL-based approaches, generating high-fitness regulatory sequences and achieving state-of-the-art cell-type specificity. Moreover, Ctrl-DNA-generated sequences capture key cell-type-specific transcription factor binding sites (TFBS), short DNA motifs recognized by regulatory proteins that control gene expression, demonstrating the biological plausibility of the generated sequences.
title Ctrl-DNA: Controllable Cell-Type-Specific Regulatory DNA Design via Constrained RL
topic Machine Learning
Artificial Intelligence
Genomics
url https://arxiv.org/abs/2505.20578