Heqin Zhu · Fenghe Tang · Quan Quan · Ke Chen · Peng Xiong · S. Kevin Zhou
RNA secondary structure plays essential roles in modeling RNA tertiary structure and further exploring the function of non-coding RNAs. Computational methods, especially deep learning methods, have demonstrated great potential and performance for RNA secondary structure prediction. However, the generalizability of deep learning models is a common unsolved issue in the situation of unseen out-of-distribution cases, which hinders the further improvement of accuracy and robustness of deep learning methods. Here we construct a base pair motif library which enumerates the complete space of locally adjacent three-neighbor base pair and records the thermodynamic energy of corresponding base pair motifs through de novo modeling of tertiary structures, and we further develop a deep learning approach for RNA secondary structure prediction, named BPfold, which employs hybrid transformer and convolutional neural network architecture and an elaborately designed base pair attention block to jointly learn representative features and relationship between RNA sequence and the energy map of base pair motif generated from the above motif library. Quantitative and qualitative experiments on sequence-wise datasets and family-wise datasets have demonstrated the great superiority of BPfold compared to other state-of-the-art approaches in both accuracy and generalizability. The significant performance of BPfold will greatly boost the development of deep learning methods for predicting RNA secondary structure and the further discovery of RNA structures and functionalities.
- Linux system
- python3.6+
- anaconda
- Clone this repo.
git clone [email protected]:heqin-zhu/BPfold.git
cd BPfold- Create and activate BPfold environment.
conda env create -f BPfold_environment.yaml
conda activate BPfold- Download model_predict.tar.gz in releases and decompress it.
wget https://github.com/heqin-zhu/BPfold/releases/download/v0.1/model_predict.tar.gz
tar -xzf model_predict.tar.gz -C paras- Download datasets BPfold_data.tar.gz in releases and decompress them.
wget https://github.com/heqin-zhu/BPfold/releases/download/v0.1/BPfold_data.tar.gz
mkdir BPfold_data && tar -xzf BPfold_data.tar.gz -C BPfold_dataFor reproduction of all the quantitative results, we provide the predicted secondary structures and model parameters of BPfold in experiments. You can directly downalod the predicted secondary structures by BPfold or use BPfold with trained parameters to predict these secondary structures, and then evaluate the predicted results.
Directly download
wget https://github.com/heqin-zhu/BPfold/releases/download/v0.1/BPfold_test_results.tar.gz
tar -xzf BPfold_test_results.tar.gzUse BPfold
- Download BPfold_reproduce.tar.gz in releases.
wget https://github.com/heqin-zhu/BPfold/releases/download/v0.1/BPfold_reproduce.pth -P paras- Use BPfold to predict test sequences.
python3 -m BPfold.main --run_name BPfold_reproduce --batch_size 32 --phase test --data_dir BPfold_data --index_name data_index.yaml --use_BPE --normalize_energy >> BPfold_reproduce.log 2>&1Evaluate
python3 -m BPfold.evaluate --data_dir BPfold_data --pred_dir BPfold_test_resultsAfter running above commands for evaluation, you will see the following outputs:
Time used: 23s
[Summary] eval_BPfold_test_results.yaml
Pred/Total num: [('PDB_test', 60, 60), ('Rfam14.5-14.10', 2048, 2048), ('archiveII', 3966, 3966), ('bpRNA', 1305, 1305)]
---------------------len>600---------------------
dataset & num & f1 & p & r \\
archiveII & 55 & 0.377 & 0.547 & 0.313\\
--------------------len<=600---------------------
dataset & num & f1 & p & r \\
PDB_test & 60 & 0.727 & 0.779 & 0.695\\
Rfam14.5-14.10 & 2048 & 0.791 & 0.777 & 0.824\\
archiveII & 3911 & 0.926 & 0.927 & 0.929\\
bpRNA & 1305 & 0.701 & 0.670 & 0.757\\
-----------------------all-----------------------
dataset & num & f1 & p & r \\
PDB_test & 60 & 0.727 & 0.779 & 0.695\\
Rfam14.5-14.10 & 2048 & 0.791 & 0.777 & 0.824\\
archiveII & 3966 & 0.918 & 0.922 & 0.921\\
bpRNA & 1305 & 0.701 & 0.670 & 0.757\\The base pair motif library is publicly available here, which contains the motif:energy pairs. The motif is represented as sequence_pairIdx_pairIdx-chainBreak where pairIdx is 0-indexed, and the energy is a reference score of statistical and physical thermodynamic energy.
For instance, CAAAAUG_0_6-3 -49.7835 represents motif CAAAAUG has a known pair C-G whose indexes are 0 and 6, with chainBreak lying at position 3.
Use BPfold to predict RNA secondary structures. The following are some examples.
python3 -m BPfold.predict --input examples/examples.fasta
python3 -m BPfold.predict --seq UUAUCUCAUCAUGAGCGGUUUCUCUCACAAACCCGCCAACCGAGCCUAAAAGCCACGGUGGUCAGUUCCGCUAAAAGGAAUGAUGUGCCUUUUAUUAGGAAAAAGUGGAACCGCCUG AGGCAGUGAUGAUGAAAAAAGAUUACCAUCAAACUUUGAGAGAUUCACAGCUCGUUGAUGCAUACUUCUUUAUAUUACCUGAGCCU
python3 -m BPfold.predict --input examples/URS0000D6831E_12908_1-117_CI0.931.bpseq --output results --save_type ctHere are the outputs after running python3 -m BPfold.predict --input examples/examples.fasta:
>> Welcome to use "BPfold" for predicting RNA secondary structure!
Loading paras/model_predict/BPfold_1-6.pth
Loading paras/model_predict/BPfold_2-6.pth
Loading paras/model_predict/BPfold_3-6.pth
Loading paras/model_predict/BPfold_4-6.pth
Loading paras/model_predict/BPfold_5-6.pth
Loading paras/model_predict/BPfold_6-6.pth
[ 1] saved in "BPfold_results/5s_Shigella-flexneri-3_CI0.980.bpseq", CI=0.980
CUGGCGGCAGUUGCGCGGUGGUCCCACCUGACCCCAUGCCGAACUCAGAAGUGAAACGCCGUAGCGCCGAUGGUAGUGUGGGGUCUCCCCAUGCGAGAGUAGGGAACUGCCAG
(((((((.....((((((((.....((((((.............))))..))....)))))).)).((.((....((((((((...))))))))....)).))...)))))))
[ 2] saved in "BPfold_results/URS0000D6831E_12908_1-117_CI0.931.bpseq", CI=0.931
UUAUCUCAUCAUGAGCGGUUUCUCUCACAAACCCGCCAACCGAGCCUAAAAGCCACGGUGGUCAGUUCCGCUAAAAGGAAUGAUGUGCCUUUUAUUAGGAAAAAGUGGAACCGCCUG
......((((((..(.(((((.......))))))(((.((((.((......))..))))))).................))))))..(((......)))..................
Finished!For more help information, please run command python3 -m BPfold.predict -h to see.
We appreciate the following open source projects:
If you use our code, please kindly consider to cite our paper:
@article {Zhu2024.10.22.619430,
author = {Zhu, Heqin and Tang, Fenghe and Quan, Quan and Chen, Ke and Xiong, Peng and Zhou, S. Kevin},
title = {Deep generalizable prediction of RNA secondary structure via base pair motif energy},
elocation-id = {2024.10.22.619430},
year = {2024},
doi = {10.1101/2024.10.22.619430},
publisher = {Cold Spring Harbor Laboratory},
URL = {https://www.biorxiv.org/content/early/2024/10/25/2024.10.22.619430},
eprint = {https://www.biorxiv.org/content/early/2024/10/25/2024.10.22.619430.full.pdf},
journal = {bioRxiv}
}