Multivariate Parametrization of Amino Acid Properties by Thin Layer Chromatography

Abstract 48 amino acids (20 coded and 28 non‐coded) were multivariately characterized by seven thin layer chromatography variables and two theoretical variables (molecular weight and side chain van der Waals volume). Principal components analysis (PCA) of the original 48 × 9 data matrix, resulted in two significant principal components which together described 88% of the variance in the data. The relationship between these principal components and the earlier estimated z‐scales for coded amino acids was investigated. In addition, it was found that the principal components had a predictive capacity similar to the z‐scales, when compared in a peptide QSAR of a series of bradykinin potentiating pentapeptides modified with only coded amino acids. Thus it is shown that TLC is a rapid technique suitable for the parametrization of amino acids.