Artificial Intelligence, Big Data and Machine Learning Approaches in Precision Medicine & Drug Discovery

Area of Science:

• Computational chemistry
• Bioinformatics
• Drug discovery

Background:

• Artificial Intelligence (AI) significantly enhances drug development by enabling rapid identification of biologically active compounds.
• Machine Learning (ML) tools and algorithms are increasingly applied across various stages of drug design and development.

Purpose of the Study:

• To provide an overview of ML applications in drug discovery.
• To highlight the role of ML in analyzing large datasets for identifying potential drug candidates and targets.

Main Methods:

• Review of ML tools (e.g., GOLD, Deep PVP, LIB SVM) and algorithms (e.g., Support Vector Machine (SVM), Random Forest (RF), Artificial Neural Network (ANN)).
• Application of ML in SNP discoveries, drug repurposing, virtual screening (LBVS, SBVS), QSAR modeling, and ADMET analysis.

Main Results:

• SVM demonstrates high performance in predicting human intestinal absorption (HIA).
• SVM and RF models successfully identified a novel colon cancer compound (JFD00950) targeting FEN1.
• ANN-based QSAR models predicted flavonoid inhibitory effects for diabetes mellitus (DM) treatment.

Conclusions:

• ML approaches are powerful tools for handling big data in drug discovery.
• ML aids in modeling small-molecule drugs, gene biomarkers, and identifying novel drug targets for various diseases.