Accurate prediction of potential druggable proteins based on genetic algorithm and Bagging-SVM ensemble classifier

Area of Science:

• Bioinformatics
• Computational Biology
• Drug Discovery

Background:

• Accurate drug target identification is crucial for new drug development.
• Machine learning offers a time- and cost-efficient alternative to traditional methods.
• Predicting druggable proteins aids in identifying potential therapeutic targets.

Purpose of the Study:

• To develop a novel machine learning-based method for predicting druggable proteins.
• To enhance the accuracy and efficiency of drug target identification.
• To provide a valuable tool for drug research and development.

Main Methods:

• Feature extraction using Chou's pseudo amino acid composition (PseAAC), dipeptide composition (DPC), and reduced sequence (RS).
• Feature selection via genetic algorithm (GA).
• Development of a prediction model using Bagging ensemble learning with a Support Vector Machine (SVM) classifier.

Main Results:

• Achieved a predictive accuracy rate of 93.78% using 5-fold cross-validation.
• The proposed method demonstrated superior performance compared to other state-of-the-art predictive methods.
• The model effectively identifies potential drug targets.

Conclusions:

• The novel GA-Bagging-SVM method shows high reference value for predicting potential drug targets.
• This approach can significantly contribute to accelerating drug research and development.
• The source code and datasets are publicly available for further research.