Advances of AI-driven Drug Design and Discovery in Pharmaceuticals - Review

Abstract

The field of drug design and discovery is undergoing a transformative shift, leveraging Artificial Intelligence (AI) and Machine Learning (ML) techniques to expedite and optimize the drug development process. Traditional methods are often costly and time-consuming, involving extensive testing and sequential stages. However, contemporary drug development integrates AI, particularly in drug identification and preclinical studies, resulting in significant resource and time savings. AI is utilized for bioactivity and physicochemical forecasting, de novo molecule design, synthesis prediction, and drug-target profile representation. This review introduces the AI-based Drug Design and Discovery System (AI-D3S), a comprehensive approach utilizing serialization, de-serialization, Particle Swarm Optimization (PSO), and Support Vector Machine (SVM). The system demonstrates superior accuracy, precision, sensitivity, and specificity compared to conventional methods, showcasing an average improvement of 8.7%. Our review study evaluates the system on two chemical databases, MAO and Biodegradation, and illustrates its efficacy in predicting drug-annotation combinations. The potential of AI in pharmaceuticals extends from drug design to personalized therapies, decision-making, and efficient resource allocation in marketing. The research envisions AI as an indispensable tool in the pharmaceutical sector, driving innovation, reducing costs, and ensuring the production of higher-quality products. The AI-D3S model presented in this study sets the stage for future advancements in drug design and discovery, offering a promising avenue for the integration of AI in revolutionizing the pharmaceutical industry.