Skip to main content
Figure 1 | BMC Bioinformatics

Figure 1

From: Survival prediction from clinico-genomic models - a comparative study

Figure 1

Breast cancer data. Results after applying the clinical model (left-hand boxplot) and the seven prediction methods to both the microarray gene expression data (center boxplot) and the combined data (right-hand boxplot). In all three boxplots, the horizontal line at zero indicates the null model with no covariate information. For the center and right-hand boxplots, the dashed magenta line indicates the median of the clinical model. Further, the green stars in the right-hand boxplot are the median of each of the seven methods when applied to the microarray gene expression data. A small value of the difference in deviance corresponds to a good prediction performance. Uni - univariate selection, PCR - principal components regression, PLS - partial least squares, SPCR - supervised PCR, SPLS - supervised PLS, and RR - ridge regression.

Back to article page