Full Description

This paper describes the first step towards building quantitative structure-propertyrelationship (QSPR) models to systematically select a group of representativemicropollutants, which will serve as a training set to develop QSPR models for watertreatment processes. A well developed optimized selection strategy was applied, whichcombined principal component analysis (PCA) and statistical experimental design. Inthis research, the initial dataset contained 183 micropollutants, mostly emergingcontaminants, selected from the peer-reviewed literature. Each compound wascharacterized by 858 molecular descriptors (i.e. these are variables used in QSPRmodeling). This resulted in a large complex multivariate dataset to which PCA wasapplied to summarize the information in the form of principal components. The firstfour principal components which captured 62.9% of the variation in the initial datasetwere used to select representative compounds using a D-optimal onion designapproach. Using this design, 22 substances were selected as structurally representativecompounds which covered the chemical domain (meaning the chemical characteristicsof all compounds) in a well-balanced manner and captured the majority of theinformation. The systematic selection approach employed here ensures that futureQSPR models are applicable to a wide range of chemicals as long as theircharacteristics fall within the original chemical domain. Includes 15 references, tables, figure.

Product Details

Edition: Vol. - No. Published: 11/01/2009 Number of Pages: 9File Size: 1 file , 770 KB