Application of genetic algorithm-Multiple linear regression for prediction of dopamine receptor 4 (D4R) antagonists of alkoxymethyl morpholines

Subject Areas :

Samira Masoomi Aladezgeh ¹ (Department of Chemistry, Faculty of Science, University of Mohaghegh Ardabili, P.O. Box 179, Ardabil. Iran)
Haniye Ghaffari Jajin ² (Department of Chemistry, Faculty of Science, University of Mohaghegh Ardabili, P.O. Box 179, Ardabil. Iran)
Eslam Pourbasheer ³ (Department of Chemistry, Faculty of Science, University of Mohaghegh Ardabili, P.O. Box 179, Ardabil, Iran)

Received: 2022-09-09 Accepted : 2022-10-18 Published : 2022-05-22

Keywords: QSAR, MLR, genetic algorithms, Alkoxymethyl morpholines,

Abstract :

In this research, by using the structural descriptors and multiple linear regression method, the quantitative structure-activity relationship studies have been carried out to predict the dopamine 4 receptors activity of, alkoxyphenylmorpholine derivatives. Appropriate descriptors were selected using the genetic algorithm method. Then a simple and strong model with a high correlation coefficient was built. The results showed that the linear techniques such as multiple linear regression coupled with a suitable variable selection method are able to provide suitable models for predicting the activity of compounds. The values of correlation coefficient (R2) and root mean square error (RMSE) for the training set were 0.729 and 0.285, respectively, and for the test set, they were 0.820 and 0.237, respectively. The presented model showed high statistical parameters that can be used to predict the activity of same compounds.