Serum SELDI-TOF MS analysis model applied to benign and malignant ovarian tumor identification

Abstract

SELDI-TOF MS serum peptide profiles of malignant and benign ovarian tumor samples were studied using a pattern recognition technique. The model of uncorrelated linear discriminant analysis (ULDA) combined with variables selection method of variance analysis was constructed to identify ovarian tumor serum samples and compared with the results obtained from principal component analysis (PCA) and partial least squares-discriminate analysis (PLS-DA). In addition, special peaks (m/z locations) as potential biomarkers were selected in this study. The good results indicate that the strategy of ULDA combined with variables selection applied to serum SELDI-TOF MS is a practicable and promising method for the ovarian malignant and benign tumor identification and selection of potential biomarkers.