Journal Title
Title of Journal: Interdiscip Sci Comput Life Sci
|
Abbravation: Interdisciplinary Sciences: Computational Life Sciences
|
Publisher
International Association of Scientists in the Interdisciplinary Areas
|
|
|
|
Authors: Mukesh C Sharma
Publish Date: 2015/07/23
Volume: 8, Issue: 1, Pages: 1-10
Abstract
Twodimensional 2D and threedimensional 3D quantitative structure–activity relationship QSAR studies were performed for correlating the chemical composition of imidazole5carboxylic acid analogs and their angiotensin II hbox AT 1 receptor antagonist activity using partial least squares and knearest neighbor respectively For comparing the three different feature selection methods of 2DQSAR knearest neighbor models were used in conjunction with simulated annealing SA genetic algorithm and stepwise coupled with partial least square PLS showed variation in biological activity The statistically significant best 2DQSAR model having good predictive ability with statistical values of r2 = 08040 and hbox pred r2 = 07764 was developed by SApartial least square with the descriptors like hbox SsCH 3count 5Chain count SdsCHEindex and Hacceptor count showing that increase in the values of these descriptors is beneficial to the activity The 3DQSAR studies were performed using the SAPLS A leaveoneout crossvalidated correlation coefficient q2=07188 and predicate activity hbox pred r2 = 07226 were obtained The information rendered by QSAR models may lead to a better understanding of structural requirements of substituted imidazole5carboxylic acid derivatives and also aid in designing novel potent antihypertensive molecules
Keywords:
.
|
Other Papers In This Journal:
|