Bioinformatics 2007 23(2):266; doi:10.1093/bioinformatics/btl576
CASPAR: a hierarchical Bayesian approach to predict survival times in cancer from gene expression data
Bioinformatics (2006) 22(12), 14951502Lars Kaderali, Thomas Zander, Ulrich Faigle, Jürgen Wolf, Joachim L. Schultze and Rainer Schrader
The authors would like to apologize for an error in Figure 2. Both plots show the same data, although the legend is correct. The corrected figure and its legend are shown below.
|
| ||||||||||||||||||||||||||||||||||||||||||||||

(t), for the groups with predicted survival >3.5 years (dashed line) and
3.5 years (solid line) on adenocarcinoma dataset, using the maximum a posteriori parameter vector
. The achieved significance level is 0.057. Right panel: True survivor functions for predicted long and short survivors (> vs