A deep learning framework for predicting molecular property based on multi-type features fusion
we propose a fusion model named DLF-MFF, which integrates four types of features from molecular fingerprints, 2D molecular graph, 3D molecular graph and molecular image to predict molecular property. Moreover, DLF-MFF is applied to identify potential anti-SARS-CoV-2 inhibitor from 2500 drugs.
Paper: [https://doi.org/10.1016/j.compbiomed.2023.107911]
Command
- Train Use train.py
Args:
data_path : The path of input CSV file. E.g. input.csv dataset_type : The type of dataset. E.g. classification or regression save_path : The path to save output model. E.g. model_save log_path : The path to record and save the result of training. E.g. log E.g.
python train.py --data_path data/test.csv --dataset_type classification --save_path model_save --log_path log
- Hyperparameters Optimization Use hyper_opti.py
Args:
data_path : The path of input CSV file. E.g. input.csv dataset_type : The type of dataset. E.g. classification or regression save_path : The path to save output model. E.g. model_save log_path : The path to record and save the result of hyperparameters optimization. E.g. log E.g.
python hyper_opti.py --data_path data/test.csv --dataset_type classification --save_path model_save --log_path log
Data We provide the three public benchmark datasets used in our study: <Data.rar>
Or you can use your own dataset:
- For training The dataset file should be a CSV file with a header line and label columns. E.g. ''' SMILES,BT-20 O(C(=O)C(=O)NCC(OC)=O)C,0 FC1=CNC(=O)NC1=O,0 ...
Citation If you find this repo useful, please cite our paper:
Mei Ma, Xiujuan Lei. A deep learning framework for predicting molecular property based on multi-type features fusion. Computers in Biology and Medicine,2024,(169):107911. https://doi.org/10.1016/j.compbiomed.2023.107911. (https://www.sciencedirect.com/science/article/pii/S0010482523013768)}
Contact If you have any question, please contact us: [email protected]