Authors:
Qingzhong Liu
1
;
Andrew H. Sung
2
;
Bernardete M. Ribeiro
3
and
Mengyu Qiao
4
Affiliations:
1
New Mexico Tech, United States
;
2
Computer Science Department, New Mexico Tech, United States
;
3
University Of Coimbra, Portugal
;
4
Institute for Complex Additive Systems Analysis, United States
Keyword(s):
Proteomics, Mass spectrometry, Manifold, Distance metric learning, Classification, Support vector machine.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Bioinformatics
;
Biomedical Engineering
;
Biomedical Signal Processing
;
Computational Intelligence
;
Soft Computing
Abstract:
Mass spectrometry becomes the most widely used measurement in proteomics research. The quality of the feature set and applied learning classifier determine the reliability of the prediction of disease status. A well-known approach is to combine peak detection and support vector machine recursive feature elimination (SVMRFE). To compare the feature selection and to search for alternative learning classifier, in this paper, we employ a distance metric learning to classification of proteomics mass spectrometry (MS) data. Experimental results show that distance metric learning is promising for the classification of proteomics data; the results are comparable to the best results by applying SVM to the SVMRFE feature sets. Results also indicate that the good potential of manifold learning for feature reduction in MS data analysis.