Abstract
Standard mixtures of proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Polynomial regression analysis was used to fit curves to the data, points obtained by plotting log10 of protein molecular weight versus electrophoretic mobility. Polynomials with orders ranging from 1 to 4 were generated. The coefficients of each equation were analyzed for statistical significance. It was found that a third order polynomial was the hightest-order equation in which all coefficients contributed significantly to the prediction of molecular weights. Using this equation, it was possible to estimate the molecular weights of known proteins in the range from 97 400 to 14 400 with a maximum error of 1 %, compared with a maximum error of 17 % when a first-order equation was used to describe the migration of the standarads.