Prediction of electrophoretic enantioseparation of aromatic amino acids/esters through MIA-QSPR

Abstract

Multivariate image analysis (MIA) descriptors have been applied to predict the enantiomer migration orders of a series of aromatic amino acids and aromatic amino esters. In MIA-QSPR, pixels of chemical structures (2D images) stand for descriptors, and structural changes account for the variance in relative migration times (RMTs). R and S enantiomers were differentiated by drawing up or down stereo bonds at the chiral carbon, and the RMT predictions of the title compounds in specific medium (20 mM Tris–citric acid background electrolyte (pH 2.50) containing 5.0 mM of (+)-18-crown-6-tetracarboxylic acid) were reliably obtained (r2 = 0.992, qLOO-CV2=0.926, and qL-20%-O-CV2=0.910) after removing two outliers from the dataset. MIA descriptors were capable to recognize the physicochemical information and may be useful to predict enantiomer migration orders of amino acids and amino esters whose pure enantiomers are unavailable.