Presentation	2008-06-26 Differential gene discovery with considering characteristic patterns Shigeyuki OBA, Shin ISHII,
PDF Download Page	PDF download Page Link
Abstract(in Japanese)	(See Japanese page)
Abstract(in English)	For better detection of significant genes with differential expression (DE) between different clinical groups based on microarray measurement, we consider, in this study, a situation where expression patterns of significant genes may show correlation which does not directly correspond to the clinical labels. In order to extract the correlation, we defined factor statistics based on a singular value decomposition of a residual matrix, that is, a gene expression profile matrix subtracted by mean differential expressions. We presented a multi-dimensional test statistic consisting of the conventional and the novel statistics, and applied the framework of empirical Bayesian statistical test. This new test had a conservative estimation of false discovery rate (FDR) and exhibited higher power than conventional statistical tests.
Keyword(in Japanese)	(See Japanese page)
Keyword(in English)	FDR / gene selection
Paper #	NLP2008-3,NC2008-13
Date of Issue

Conference Information
Committee	NC
Conference Date	2008/6/19(1days)
Place (in Japanese)	(See Japanese page)
Place (in English)
Topics (in Japanese)	(See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To	Neurocomputing (NC)
Language	JPN
Title (in Japanese)	(See Japanese page)
Sub Title (in Japanese)	(See Japanese page)
Title (in English)	Differential gene discovery with considering characteristic patterns
Sub Title (in English)
Keyword(1)	FDR
Keyword(2)	gene selection
1st Author's Name	Shigeyuki OBA
1st Author's Affiliation	Graduate School of Informatics, Kyoto University()
2nd Author's Name	Shin ISHII
2nd Author's Affiliation	Graduate School of Informatics, Kyoto University
Date	2008-06-26
Paper #	NLP2008-3,NC2008-13
Volume (vol)	vol.108
Number (no)	101
Page	pp.pp.-
#Pages	6
Date of Issue