Feature Extraction and Classification of Proteomics Data Using Stationary Wavelet Transform and Naive Bayes Classifier

The purpose of the current study was to investigate the changes of serum proteome and to discover potential biomarkers from a publicly available proteomic ovarian dataset. A workflow that combines stationary wavelet transform with naive Bayes classifier was presented to select candidate biomarkers form 253 proteomic serum profiles of cancer and control. The method identified correlative mass points and obtained a discriminative pattern with 96.7% sensitivity and 92.7% specificity.