A L1-regularized feature selection method for local dimension reduction on microarray data. (April 2017)

• Record Type: Journal Article
• Authors: Guo, Shun; Guo, Donghui; Chen, Lifei; Jiang, Qingshan
• Is Part Of: Computational biology and chemistry. Volume 67(2017)
• Page Start: 92
• Page End: 101
• Publication Date: 2017-04
• Subjects: Local dimension reduction -- Classification -- L1-regularized logistic regression -- Microarray data -- Partial least squares (PLS); Chemistry -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; Biochemistry -- Data processing; Biology -- Data processing; Molecular biology -- Data processing; Periodicals; Electronic journals; 542.85
• DOI: 10.1016/j.compbiolchem.2016.12.010
• Languages: English
• ISSNs: 1476-9271

Abstract

Graphical abstract: Highlights: A two-stage local dimension reduction method for microarray data is proposed. A new L1-regularized feature selection method is defined in first stage and it can obtain the global optimal solution. Extensive experiments show the effectiveness of the proposed method. The mining results can be used for subtype prediction. Abstract: Dimension reduction is a crucial technique in machine learning and data mining, which is widely used in areas of medicine, bioinformatics and genetics. In this paper, we propose a two-stage local dimension reduction approach for classification on microarray data. In first stage, a new L1-regularized feature selection method is defined to remove irrelevant and redundant features and to select the important features (biomarkers). In the next stage, PLS-based feature extraction is implemented on the selected features to extract synthesis features that best reflect discriminating characteristics for classification. The suitability of the proposal is demonstrated in an empirical study done with ten widely used microarray datasets, and the results show its effectiveness and competitiveness compared with four state-of-the-art methods. The experimental results on St Jude dataset shows that our method can be effectively applied to microarray data analysis for subtype prediction and the discovery of gene coexpression.