QSAR Model Study of 2,3,4,5-tetrahydro-1H-pyrido[4,3-b]indole of Cystic- brosis-transmembrane Conductance-regulator Gene Potentiators

Background: Cystic fibrosis (CF) is a genetic disease, which has no effective treatment. Objective: The aim of this study is to predict the EC50 value of 2,3,4,5-tetrahydro-1H-pyrido[4,3-b]indole core as a novel chemotype of potentiators to establish a highly predicting quantitative structure-activity relationship model. Methods: 41 products were optimized, and a linear model was built by a heuristic method in CODESSA program. In this study, 3 descriptors were selected and utilized to build a nonlinear model in gene expression programming. Results: The square of the correlation coefficient of the heuristic method is 0.57, and the s2 is 0.30. In gene expression programming, the square of correlation coefficient and the mean square error for the training set are 0.74 and 0.13, respectively. The square of correlation coefficient and the mean square error for the test set are 0.70 and 0.27, respectively. Conclusion: The GEP model has stronger predictive ability to help develop the novel structure of 2,3,4,5-tetrahydro-1H-pyrido[4,3-b]indole of cystic-brosis-transmembrane conductance-regulator gene potentiators.