COMPARATIVE PREDICTION OF THE RETENTION BEHAVIOR OF SMALL PEPTIDES INSEVERAL REVERSED-PHASE HIGH-PERFORMANCE LIQUID-CHROMATOGRAPHY COLUMNSBY USING PARTIAL LEAST-SQUARES AND MULTIPLE LINEAR-REGRESSION

The retention behaviour of small standard peptides in reversed-phase h igh-performance liquid chromatography (RP-HPLC) is evaluated. Chromato graphic retention of small peptides highly correlates with the summing of the hydrophobic contributions of their respective amino acid resid ues. Prediction of peptide retention time is accomplished in four diff erent RP columns (Ultrasphere Octyl, Ultrasphere ODS, Polymeric revers ed phase PLRP-S and Nova-pak C-18) under the same chromatographic cond itions. The elution profiles of a mixture of 25 small standard peptide s are highly correlated among the four RP columns. In addition, RP col umns exhibit similar behaviour as far as the pattern of amino acid con tribution to the retention of small peptides is concerned. Comparative prediction of peptide retention times is carried out using Multiple L inear Regression (MLR) and Partial Least Squares (PLS) regression. PLS regression provides better prediction ability than MLR. Following the PLS procedure, a reasonable estimation of peptide retention times, af ter calculating the corresponding amino acid retention coefficients, i s possible.