Quantitative structure-property relationship study of acidity constants ofsome 9,10-anthraquinone derivatives using multiple linear regression and partial least-squares procedures
M. Shamsipur et al., Quantitative structure-property relationship study of acidity constants ofsome 9,10-anthraquinone derivatives using multiple linear regression and partial least-squares procedures, TALANTA, 54(6), 2001, pp. 1113-1120
A quantitative structure-property relationship study is suggested for the p
rediction of acidity constants of some recently synthesized 9,10-anthraquin
one derivatives in binary methanol-water mixtures. Modeling of the acidity
constant of the anthraquinones as a function of physicochemical parameters
and mole fraction of methanol was established by means of the partial least
-squares algorithm based on singular value decomposition (PLS-SVD) and mult
iple linear regression. The PLS-SVD procedure resulted in a better predicti
on ability and was found to be insensitive to noneffective descriptors. The
classification of anthraquinones by the calculated descriptors was establi
shed. (C) 2001 Elsevier Science B.V. All rights reserved.