Exploring Novel QSRRs for Simulation of Gas Chromatographic Retention Indices of Diverse Sets of Terpenoids inPistacia LentiscusL. Essential Oil Using Stepwise and Genetic Algorithm Multiple Linear Regressions

Abstract

The present work describes convenient and interpretable models by step-wise multiple linear regression (SW-MLR) and genetic algorithm-multiple linear regression (GA-MLR). These quantitative structure-retention relationship (QSRR) based strategies have been successfully used to predict the retention indices (RIs) of a series of natural compounds found in the essential oil of Pistacia lentiscus L. The dataset was divided into training set (51 compounds) and test set (25 compounds), randomly. The prediction capabilities of both approaches have been appraised by their exertion to test set compounds. The fitness statistics of the models were assessed to be satisfactory which resulted in accurate predictions. The best optimized models for SW-MLR and GA-MLR consisted 5 (X0, Qtot, Dv, HATS3e and MATS4e) and 6 (MATS2e, DELS, ATS4e, PW5, PCD and W) molecular descriptors, respectively. Our investigation revealed the superiority of SW-MLR (R 2 =0.96, Q 2 LOO =0.95 and Q 2 LGO =0.94 for training set; REP=3.7 for test set) standpoint against the best GA-MLR model (R 2 =0.955, Q 2 LOO =0.937 and Q 2 LGO =0.934 for training set; REP=7.5 for test set) for estimation of RIs of similar or unknown compounds.

A quantitative structure–activity relationship study on CXL017 derivatives as effective drugs for cancer treatment

“…However, for a better understanding of the inhibitory mechanism of a broad set of chemical drugs, some complimentary criteria or equations should also be considered like adjusted R 2 ( R 2 adj ) according to the following formula: where p and n , respectively, denote the number of predictors and/or molecular descriptors as well as the number of molecules in the training set. [ 55,59 ]…”

“…where p and n, respectively, denote the number of predictors and/or molecular descriptors as well as the number of molecules in the training set. [55,59] The SE of the estimate(s) is also an important statistical feature explaining the scattering pattern of the predicted values around the regression line. The less SE of the estimate, the higher predictiveness of the generated model is.…”

“…Otherwise, it will be discarded if the corresponding probability of F variables is large. [55] As it is customary in such sorts of studies, some statistical parameters are usually used to judge the quality of the generated models. These statistical terms include correlation coefficient (R), determination coefficient (R 2 ), adjusted R 2 (R 2 adj ), among others.…”

Predictionof Environmental FateandToxicityofInsecticidesUsing Multi‐Target QSAR Approach

“…The most promising technique for in silico estimation of the values of the property/activity of a compound is the Quantitative Structure-Activity Relationship (QSAR) model. [23][24][25][26] It is a statistical data analytical procedure in which quantitative endpoints of chemicals are correlated with mathematical descriptors encoding the molecular structure of the compounds. In the field of environmental research, the application of QSAR models to prioritize, screen and rank chemicals is well established.…”

“…These methods, which are based on computer tools, are being used by both regulatory agencies and companies as a prediction model to fill the existing data gaps. The most promising technique for in silico estimation of the values of the property/activity of a compound is the Quantitative Structure‐Activity Relationship (QSAR) model [23–26] . It is a statistical data analytical procedure in which quantitative endpoints of chemicals are correlated with mathematical descriptors encoding the molecular structure of the compounds.…”