Skip to main content
. Author manuscript; available in PMC: 2009 Dec 15.
Published in final edited form as: Clin Cancer Res. 2008 Dec 15;14(24):8213–8220. doi: 10.1158/1078-0432.CCR-08-0095

Figure 2.

Figure 2

Validation of the 35-gene prognostic signature in Raponi’s cohort (n = 129). The data was randomly partitioned into a training set (n = 65) and a test set (n = 64). In the training set, each tumor sample was classified as low-risk if the correlation with the good prognostic group in Beer’s cohort is greater than −0.15; otherwise, it is classified as high-risk. The same cutoff was applied to the test set in patient stratification.