Premium
Quantitative Structure–Activity Relationship (QSAR) Modeling of Human Blood : Air Partitioning with Proper Statistical Methods and Validation
Author(s) -
Basak Subhash C.,
Mills Denise,
Hawkins Douglas M.,
Kraker Jessica J.
Publication year - 2009
Publication title -
chemistry and biodiversity
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.427
H-Index - 70
eISSN - 1612-1880
pISSN - 1612-1872
DOI - 10.1002/cbdv.200800111
Subject(s) - quantitative structure–activity relationship , chemistry , partial least squares regression , molecular descriptor , partition coefficient , linear regression , regression analysis , stepwise regression , biological system , statistics , mathematics , stereochemistry , chromatography , biology
Blood : air partition coefficient (BA pc ) is important in assessing toxicokinetics of chemicals. Since very few experimental data are available, quantitative structure–activity relationship (QSAR) models based on calculated molecular descriptors can be useful in estimating BA pc . Since all descriptors used in the analysis are computed strictly from structure, they can be applied to any chemical, real or hypothetical. In this article, we report models for BA pc estimation using four methods, including stepwise ordinary least‐squares regression, which is commonly used in QSAR studies but often results in an inflated ‘naïve’ q 2 , over‐representing the predictive ability of the model. The models developed using proper statistical techniques had q 2 values of 0.825 and 0.841, and may be used to reliably predict BA pc values for new compounds that are structurally similar to those upon which the models are based. The models developed using improper techniques had associated q 2 values, as computed using naïve methods, of 0.920 and 0.934, severely overstating their actual quality.