Buiilding QSAR model for dipeptidyl peptidase IV inhibitors using linear regression

Abstract

Big data brings challenges and opportunities for research in health science in general. In this era, computer becomes a powerful lab for drug development. Specifically, in this thesis, we built a Quantitative Structure Activity Relationship (QSAR) model based on a linear regression algorithm with input was the BindingDB database to predict the IC50 value in DPPIV inhibitors against Diabetes. Such kind of work reduce cost and save a lot of time for drug development as the DPPIV keep an important role in Diabetes Mellitus disease. The model including 6 features with R-squared was 0.840 giving the best performance after testing from 1 to 30 features