Tertiary Structure-based RNA Inverse Folding with Generative Diffusion Models
Please refer to requirements.txt for the required packages.
Model checkpoint can be downloaded from here. Another checkpoint trained on the full dataset (with extra 0.1 Gaussian noise for coordinates) can be downloaded from here.
Download and put the checkpoint files in the ckpts folder.
Inference demo notebook to get started:
Run the following command to run the example for one sequence generation:
CUDA_VISIBLE_DEVICES=0 python main.py --PDB_file example/R1107.pdbThe generated sequence will be saved in exp_inf/fasta/R1107_0.fasta.
Multiple sequence generation can be run by:
CUDA_VISIBLE_DEVICES=0 python main.py --PDB_file example/R1107.pdb --config.eval.n_samples 10Adjusting the conditional scaling weight can be done by:
CUDA_VISIBLE_DEVICES=0 python main.py --PDB_file example/R1107.pdb --config.eval.n_samples 10 --config.eval.dynamic_threshold --config.eval.cond_scale 0.4If you find this work useful, please cite:
@article{huang2024ribodiffusion,
title={RiboDiffusion: Tertiary Structure-based RNA Inverse Folding with Generative Diffusion Models},
author={Han Huang and Ziqian Lin and Dongchen He and Liang Hong and Yu Li},
journal={arXiv preprint arXiv:2404.11199},
year={2024}
}
This project is licensed under the MIT License.