**Figure 1 : Procedures to generate permutated data.**

Let *x _{ijk}* be the smoothed values of CGH data (logR) for the genomic location

for stage

investigate difference of

to the smallest mean. Then each difference between two different stages takes into account by S

The later formula detect in addition to the difference between two stages. To avoid the variability and standardization of

each stage

generates

same amount of samples for each stage and redefine the data as x

generate the

Solvang et al.*Journal of Medical Statistics and Informatics* 2014 **2**:10DOI : 10.7243/2053-7662-2-10