Predicting drug-target interactions based on fusing multiple features with data balancing and feature selection techniques Predicting drug-target interaction (DTI) is an important research area in the field of drug discovery. This framework proposes a novel drug–target interaction prediction method called SRX-DTI. First, we extract various descriptors from the protein sequences; and the drug is encoded as an FP2 molecular fingerprint. Besides, we present the One-SVM-US technique to deal with imbalanced data. We also developed the FFS-RF algorithm to remove the irrelevant features to obtain the best optimal features. Finally, the balanced dataset with optimal features is given to the XGBoost classifier to identify DTIs.
In this research, four golden standard datasets, including enzymes (EN), G-protein-coupled receptors (GPCR), ion channel (IC), and nuclear receptors (NR) released by Yamanishi et al. All these datasets are freely available from http://web.kuicr.kyoto-u.ac.jp/supp/yoshi/drugtarget/.
-
Python version: '3.9' or higher
-
You have to install the required libraries
- Run ./feature extraction/00-AAC.py: extract AAC descriptor (for other descriptors, just change to related Python code).
- Run ./NR-run/run.py: make a balanced dataset and feature selection.
- Run ./DTI prediction/DTI_predictor.py: predict drug-target interactions, and evaluate the results with five cross-validations.
If you use our source code, dataset, and experiments for your research or development, please cite our paper: Khojasteh, H., Pirgazi, J., & Ghanbari Sorkhi, A. (2023). Improving prediction of drug-target interactions based on fusing multiple features with data balancing and feature selection techniques. Plos one, 18(8), e0288173. [https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0288173]
If you have any questions, do not hesitate to contact me at [email protected]
, I will be happy to assist.