On the Rational Selection of Test Series. I. Principal Component Analysis Combined with Multidimensional Mapping (PCMM Method)

Abstract

A procedure which is a combination of 2 basic steps, principal component method and multidimensional mapping (PCMM) is described. The method minimizes collinearities and still yields test series of sufficient high data variance. The variance coeff. VS and the determinant of the correlation matrix DS were used. The method was applied to a series of monosubstitutions and the results compared well with other QSAR methods as tools to rationalize drug design.