Computational Drug Repurposing: Exploring New Uses for Existing Molecules in the Treatment of Diabetes Mellitus

The computational method of drug discovery is making remarkable, revolutionary changes in the identification and development of new drugs. Drugs already on the market with a claimed therapeutic application are repurposed for other major applications through computational drug discovery, a cost-effective, time-saving approach. By identifying the disease pathway and the pharmacological profile of the drug under investigation, drug repurposing can be performed on MATLAB rather than in a wet lab. For example, a Graph Neural Network (GNN) employs embeddings of twodimensional drug molecular structures, thereby improving repurposing models. Diabetes is one of the most serious disorders; with so many drugs claiming to regulate blood glucose levels, computational methods can repurpose the other existing drugs with more therapeutic effects. Using docking, the affinity of the agents can be predicted and thus repurposed for new applications. Rather than starting from scratch by discovering a new moiety to treat a disease that consumes significant time, it is worthwhile to repurpose an existing drug. By doing so, the drug discovery timeline is shortened, thereby improving patient compliance. Computational modeling forms the basis for drug discovery, helping overcome time constraints and reducing the need for large numbers of animal and human trials. This review highlights the computational repurposing of existing drugs for the treatment of diabetes mellitus, providing a thorough discussion of the challenges and opportunities involved.