KailiWang1 / RLBind

a deep learning architecture for RNA-ligand binding sites prediction
Apache License 2.0
16 stars 3 forks source link

About RLBind

A deep convolutional neural network-based model by integrating local and global features, sequence and structure properties is constructed to predict RNA-ligand binding sites.

The benchmark datasets can be found in ./data_cache/, the codes for RLBind are available in ./src. And the results and model are saved in ./results. Furthermore, the demo and corresponding documentation files can be found in ./demo. See our paper for more details.

[Paper:] Wang K, Zhou R, Wu Y, et al. RLBind: a deep learning method to predict RNA–ligand binding sites. Briefings in Bioinformatics, 2023, 24(1), bbac486.

Requirements

The easiest way to install the required packages is to create environment with GPU-enabled version:

conda env create -f environment_gpu.yml
conda activate RLBind_env

Testing the model

cd ./src/
python predict.py

Re-training your own model for the new dataset

cd ./src/
python training.py

contact

Kaili Wang: kailiwang@dhu.edu.cn