Machine Learning and Artificial Intelligence: A Paradigm Shift in Big Data-Driven Drug Design and Discovery

Abstract

Background: The lengthy and expensive process of developing a novel medicine often takes many years and entails a significant financial burden due to its poor success rate. Furthermore, the processing and analysis of quickly expanding massive data necessitate the use of cutting-edge methodologies. As a result, Artificial Intelligence-driven methods that have been shown to improve the efficiency and accuracy of drug discovery have grown in favor.

Objective: The goal of this thorough analysis is to provide an overview of the drug discovery and development timeline, various approaches to drug design, and the use of Artificial Intelligence in many aspects of drug discovery.

Methods: Traditional drug development approaches and their disadvantages have been explored in this paper, followed by an introduction to AI-based technology. Also, advanced methods used in Machine Learning and Deep Learning are examined in detail. A few examples of big data research that has transformed the field of medication discovery have also been presented. Also covered are the many databases, toolkits, and software available for constructing Artificial Intelligence/Machine Learning models, as well as some standard model evaluation parameters. Finally, recent advances and uses of Machine Learning and Deep Learning in drug discovery are thoroughly examined, along with their limitations and future potential.

Conclusion: Artificial Intelligence-based technologies enhance decision-making by utilizing the abundantly available high-quality data, thereby reducing the time and cost involved in the process. We anticipate that this review would be useful to researchers interested in Artificial Intelligencebased drug development.

Keywords: Artificial intelligence; Big data; Computer-aided drug design; Deep learning; Drug discovery; Machine learning; Medicinal chemistry; Quantitative structure-activity relationship.