Computational Approaches for Structure-Assisted Drug Discovery and Repurposing

The drug discovery process is highly challenging due to the excess cost, time, and resource requirements. From identifying potential molecules to their utilization in clinics as drugs, molecules face a variety of validations and many counterparts land up to rejection in between the process. Many potent molecules seem effective in treating the pathologies of the disease but are discarded due to their toxic nature in in-vitro and in-vivo conditions. Repurposing of known already available drugs can provide strong candidates that have already gone through toxicity tests and will thus reduce the time for the drug to be implemented as a treatment for a new disease. Structural information about the disease target and the potential lead molecules facilitates the identification of significant interactions that can convey regulation of the target by inhibiting it. Computational tools and approaches have immensely supported and accelerated the process of drug discovery by providing effective predictions based on structural information. The lead computational approach comprising identification of pharmacophores, physiochemical and biological activities, drug target interaction and binding affinity, and analysis of the binding stability, together lead to the identification of a potent druglike molecule. Further, different AI/ML-based models have come forth to assist drug design on the basis of structure and to increase the pace of the process without compromising accuracy. Therefore, this piece of work elaborates on the significance of different computational approaches, and recent advancements with a lead focus on the software, tools, and models developed for use in drug discovery.