A Wrapper Feature Selection Method Based on Simulated Annealing Algorithm for Prostate Protein Mass Spectrometry Data

Protein mass spectrometry is an integration of mass spectrometry and biological chip techniques, and it shows great potential for exploration of biomarkers and diagnosis of diseases. But the curse of dimensionality inherently from mass spectrometry data makes the dimensionality reduction a necessary phase of proteomic pattern recognition before classification. This paper presents a simulated annealing algorithm to select discriminant feature subsets. Experiments indicate that this wrapper feature selection method performs well and outperforms the other reported methods.