Hierarchical modeling of alternative exon usage associations with survival

Exon expression platforms have allowed the detection of associations between alternative exon usage (AEU) and the proliferation of malignant cells in cancer. However, due to inadequate number of studies performed on AEU and the approaches utilized to detect AEU events, well established biomarkers for GBM are not available. The expression of exons corresponding to 25,403 genes was related to the survival of 328 patients diagnosed with Glioblastoma multiforme (GBM). An approach that takes exon expression into account was adopted to detect the association between exon expression and survival. Association between expression and survival were identified in 22 single-exon genes 248 genes with 2-25 exons, 1430 genes with >;24 and <;50 exons and 215 genes with >;50 exons. Among the multiple-exon genes exhibiting AEU were epidermal growth factor (EGF) and nidogen2 (NID2) that have known association with GBM. These results are consistent with reports that these genes have 12 and 10 transcripts, respectively.