Analysis of Proteomic Mass Spectra Based on Multivariate Feature Fusion Visualization

Abstract

Protein mass spectrometry (MS) pattern recognition has recently emerged as a new method for cancer diagnosis. Application of proteomic mass spectra coupled with pattern classification techniques to discover novel biomarkers has been successfully used for the predictive diagnoses of several cancer diseases. However, the extraction of good features that can represent the identities of different classes plays the frontal critical factor for effective classification. In addition, another major problem is how to effectively handle a large number of features. This paper address these two frontal issues for MS classification. We propose a method based on graphical multivariate feature fusion and use it to offer a visual representation of high dimensional data. The graphical processing method we propose, relies on using a multilayered structure of feature fusion which produces as output of the lower dimensional representation. We implement feature fusion by combining method of feature selection and feature extraction. The proposed methodology was tested using public MS-based cancer datasets and the results are promising.