The Emergence of In-Silico Models in Drug Target Interaction System: A Comprehensive Review

Abstract

ABSTRACT: The dawn of computational models in healthcare has revolutionised the drug development industry. The wet lab experiments entail enormously expensive and laborious procedures. As a result, the applications of computational designs have been a better replacement for manual experimentations. Identifying drug-target interaction (DTI) is a vital drug design process. In this review, we have explored the various computational methodologies actively used in the field of DTI prediction. We have hierarchically categorised the models into three broad domains: ligand-based, structure-based and chemogenic. We have further classified the domains into their subcategories. The functioning and latest developments achieved in each subcategory are further analysed in depth. This review offers a comprehensive overview of the tools and methodologies of each model. We have also compared the advantages and limitations of each model in every category. Finally, we look into the future scope of the machine learning models by addressing the possible difficulties faced in DTI. This article serves as an insight into the various models used in DTI prediction.