https://github.com/bowang-lab/ctrl-dna
Ctrl-DNA: Controllable Cell-Type-Specific Regulatory DNA Design via Constrained RL
Science Score: 36.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
✓.zenodo.json file
Found .zenodo.json file -
○DOI references
-
✓Academic publication links
Links to: arxiv.org -
○Committers with academic emails
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (14.2%) to scientific vocabulary
Repository
Ctrl-DNA: Controllable Cell-Type-Specific Regulatory DNA Design via Constrained RL
Basic Info
Statistics
- Stars: 7
- Watchers: 2
- Forks: 0
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
Ctrl-DNA: Controllable Cell-Type-Specific Regulatory DNA Design via Constrained RL
Overview
We present Ctrl-DNA, a constrained reinforcement learning framework for the controllable design of cell-type-specific regulatory DNA sequences. Ctrl-DNA fine-tunes autoregressive genomic language models by framing sequence generation as a biologically informed constrained optimization problem. Using a value-model free, Lagrangian-guided policy optimization strategy, Ctrl-DNA iteratively refines sequences to maximize gene expression in a target cell type while suppressing activity in off-target cell types. Applied to human enhancer and promoter datasets, Ctrl-DNA generates biologically plausible, high-fitness sequences enriched for key transcription factor motifs, achieving state-of-the-art specificity and performance in regulatory sequence design.
Installation
Clone the repository and install the required dependencies:
bash
git clone https://github.com/bowang-lab/Ctrl-DNA.git
cd ctrl-dna
pip install -r requirements.txt
Data Preprocessing
We follow the preprocessing pipeline from Genentech/regLM. Please refer to their repository for detailed instructions.
Training
To train the model on the enhancer and promoter dataset using our method, run:
bash
bash reinforce_lagrange_promoters.sh
bash reinforce_lagrange_enhancers.sh
Acknowledgements
Our implementation builds upon several open-source projects:
- regLM: Provided the reward model architecture and data preprocessing pipeline.
- TACO: Supplied the reinforcement learning framework that our method extends.
We sincerely thank the authors of these projects for making their code and datasets publicly available.
Citation
If you find this work useful, please cite our paper:
bibtex
@misc{chen2025ctrldnacontrollablecelltypespecificregulatory,
title={Ctrl-DNA: Controllable Cell-Type-Specific Regulatory DNA Design via Constrained RL},
author={Xingyu Chen and Shihao Ma and Runsheng Lin and Jiecong Lin and Bo Wang},
year={2025},
eprint={2505.20578},
archivePrefix={arXiv},
primaryClass={cs.LG},
url={https://arxiv.org/abs/2505.20578},
}
Owner
- Name: WangLab @ U of T
- Login: bowang-lab
- Kind: organization
- Location: 190 Elizabeth St, Toronto, ON M5G 2C4 Canada
- Website: https://wanglab.ml
- Repositories: 11
- Profile: https://github.com/bowang-lab
BoWang's Lab at University of Toronto
GitHub Events
Total
- Issues event: 1
- Watch event: 5
- Push event: 1
- Public event: 1
Last Year
- Issues event: 1
- Watch event: 5
- Push event: 1
- Public event: 1
Committers
Last synced: about 1 year ago
Top Committers
| Name | Commits | |
|---|---|---|
| xingyu617 | e****y@g****m | 9 |
| Rex Ma | r****2@g****m | 3 |
| Lawrence Lin | l****n@g****m | 1 |
Issues and Pull Requests
Last synced: about 1 year ago
All Time
- Total issues: 0
- Total pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Total issue authors: 0
- Total pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 0
- Pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Issue authors: 0
- Pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Top Authors
Issue Authors
- bilemond (1)
Pull Request Authors
Top Labels
Issue Labels
Pull Request Labels
Dependencies
- bpnet-lite <0.6.0
- captum ==0.5.0
- enformer_pytorch <0.8.7
- numpy *
- pandas *
- pytest *
- pytest-cov *
- pytorch-lightning <2.0
- setuptools *
- torch <2.0
- torchmetrics *