Abstract

In this paper, we introduce a multivariate approach for selecting genes for predicting survival outcomes of cancer patients in gene expression microarray studies. Combined with survival analysis for gene filtering, the method makes full use of individual’s survival information (both censored and uncensored) in selecting informative genes for survival outcome prediction. Application of our method to published data on epithelial ovarian cancer has identified genes that discriminate unfavorable and favorable outcomes with high significance ( χ2 = 21.933, p = 3e - 06 ). The method can also be generalized to categorical variables for selecting gene expression signatures for predicting tumor metastasis or tumor subtypes.

OriginalsprogEngelsk
TitelAdvances in Computer and Information Sciences and Engineering
RedaktørerTarek Sobh
ForlagSpringer Science+Business Media
Publikationsdato2008
Sider405-409
ISBN (Trykt)978-1-4020-8740-0
DOI
StatusUdgivet - 2008

Fingeraftryk

Dyk ned i forskningsemnerne om 'Gene selection for predicting survival outcomes of cancer patients in microarray studies'. Sammen danner de et unikt fingeraftryk.

Citationsformater