PREDICTING DRUG-DRUG INTERACTION USING MACHINE LEARNING

Abstract

The ever-expanding landscape of pharmaceutical compounds and medical treatments has underscored the critical need for an advanced system to predict drug-drug interactions (DDIs) using machine learning. This paper delves into the development of a novel predictive model that leverages diverse data sources, including chemical structures, pharmacological properties, and scientific literature, to comprehensively analyse and foresee potential DDIs. The current state of DDIs is rife with limitations in scope and accuracy, making the development of an advanced system not only a necessity for enhancing patient safety but also a boon to pharmaceutical research and regulatory oversight. The relationship between two medications in which one drug’s pharmacological actions are altered by another is known as a drug-drug interaction (DDI). Favourable drug-drug interactions (DDIs) generally contribute to improved therapeutic results for patients. On the contrary, adverse DDIs pose a significant risk, often resulting in undesirable drug reactions that can potentially lead to the withdrawal of a drug from the market and even patient fatalities. Consequently, the identification of DDIs is essential for advancing new drug development and effectively managing diseases. Keywords: Drug, Drug-drug interaction, Machine learning, Bioinformatics, Drug development.