Biomarker discovery using 1-norm regularization for multiclass earthworm microarray gene expression data

详细信息	查看全文 | 推荐本文 |

• 作者：Xiaofei Nan^a ; ^{xiaofei.nan@gmail.com} ; ^{xnan@olemiss.edu}  ; [Author Vitae] ; Nan Wang^b ; ^{nan.wang@usm.edu}  ; [Author Vitae] ; Ping Gong^c ; ^{Ping.Gong@usace.army.mil}  ; [Author Vitae] ; Chaoyang Zhang^b ; ^{Chaoyang.zhang@usm.edu}  ; [Author Vitae] ; Yixin Chen^a ; ^{ychen@cs.olemiss.edu}  ; [Author Vitae] ; Dawn Wilkins^a ; ^{dwilkins@cs.olemiss.edu}  ; [Author Vitae]
• 关键词：Biomarker discovery ; 1-norm regularization ; Multiclass classification ; Microarray gene expression data
• 刊名：Neurocomputing
• 出版年：2012
• 期刊代码：95_09252312
• 类别：cp
• 出版时间：1 September, 2012
• 卷：92
• 期：Complete
• 页码：36-43
• 文件大小：341 K

摘要

Novel biomarkers can be discovered through mining high dimensional microarray datasets using machine learning techniques. Here we propose a novel recursive gene selection method which can handle the multiclass setting effectively and efficiently. The selection is performed iteratively. In each iteration, a linear multiclass classifier is trained using 1-norm regularization, which leads to sparse weight vectors, i.e., many feature weights are exactly zero. Those zero-weight features are eliminated in the next iteration. The empirical results demonstrate that the selected features (genes) have very competitive discriminative power. In addition, the selection process has fast rate of convergence.