Premium
A Novel Quantitative Structure‐Biodegradability Relationship (QSBR) of Substituted Benzenes Based on MHDV Descriptor
Author(s) -
Liu Yan,
Liu ShuShen,
Cui ShiHai,
Cai ShaoXi
Publication year - 2003
Publication title -
journal of the chinese chemical society
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.329
H-Index - 45
eISSN - 2192-6549
pISSN - 0009-4536
DOI - 10.1002/jccs.200300047
Subject(s) - correlation coefficient , chemistry , mean squared error , molecular descriptor , linear regression , statistics , coefficient of determination , root mean square , biological system , quantitative structure–activity relationship , mathematics , stereochemistry , electrical engineering , biology , engineering
Abstract The molecular holographic distance vector (MHDV) is employed to characterize the structures of 51 substituted benzenes. 29 descriptors from 91 MHDV ones have nonzero values where 3 descriptors have only one nonzero sample and 1 descriptor only two nonzero samples. A genetic algorithm is used to select an optimal combination of the variables from the remaining 25 nonzero descriptors. Then the optimal descriptors are employed to relate to the relative biodegradability using multiple linear regression method. The 6‐variable linear model developed has high quality where the correlation coefficient of estimations and the root mean square error of estimations are 0.9604 and 0.280, respectively, and the correlation coefficient of predictions and the root mean square error of predictions for leave‐one‐out procedure are 0.9471 and 0.324, respectively.