From roots to codes: Applications of computer-aided drug discovery from medicinal plants

Abstract

Medicinal plants are enriched sources of drugs due to their chemically and functionally diverse secondary metabolites. Conventional methods of screening these bioactive metabolites require a long time, incur high costs, utilize substantial amounts of solvents, involve repeated exposure to heat (harmful for the thermo-labile compounds), and often lead to undesirable products. However, computer-aided drug discovery (CADD) has emerged as a versatile and cost-effective tool for expediting the screening of plant-based bioactive compounds. In this review, we focus on structure-based and ligand-based drug discovery approaches and their roles in rationalizing drug development. We discuss the technicalities, limitations, and successful applications of key components of structure-based (molecular docking and molecular dynamics) and ligand-based (quantitative structure–activity relationship and pharmacophore modeling) approaches in medicinal plant-derived drug discovery. Furthermore, our study highlights the significance of the pharmacokinetic and ADMET profiling of plant-based therapeutics in drug development. Besides the applications of computational drug discovery, we also briefly overview the corroborative in-silico and experimental approaches to identify potential therapeutics from medicinal plants. This study suggests that integrating advanced Machine Learning and Artificial Intelligence approaches could alleviate the challenges faced by traditional computational techniques, thus enriching the screening process.