Figure 3

Clustering Analysis of 56 genetic biomarkers for test set FH2. Test set FH2 was standardized (subtracted by its mean and divided by its standard deviation) for each gene and a hierarchical clustering analysis was performed on this set. McQuitty's or WPGMA method, which uses the average distance between clusters weighted by uneven cluster sizes, was used. Three functional subclusters of the 56 biomarkers were identified by examining common biological functions of these gene subclusters by the function annotation tool in DAVID database (http://david.abcc.ncifcrf.gov).