AKT subgroups have distinct genomic alterations.
(A) Copy number alterations in TCGA AKT subgroups. The GISTIC method was applied to TCGA samples in each subgroup with copy number information. Data are presented as a G score which is an integrated score of the prevalence of the copy-number change times the average (log2-transformed) amplitude. The green line shows significance threshold (FDR q values to account for multiple-hypothesis testing). Regions with subgroup-specific CNA are highlighted in yellow. (B) Distribution of clinical information and mutations, CNA and mRNA expression for glioma-associated genes in AKT subgroups. The 218 TCGA GBM cases with gene expression, consensus putative copy number alteration and validated mutation data ,  was used for this analysis (The cBio Cancer Genomics Portal; http://www.cbioportal.org/). Gene expression is represented as z scores calculated relative to diploid tumors for each gene and are the median value of 3 mRNA platforms (Affymetrix U133A and Exon arrays and Agilent custom array). There was a statistically significant enrichment of IDH1 mutations in the SL and EGFR and CDKN2A mutations plus CNA in the CLAS subtype (p < 0.02).