BackgroundChanges in promoter DNA methylation pattern of genes involved in key biological pathways have been reported in glioblastoma. Genome-wide assessments of DNA methylation levels are now required to decipher the epigenetic events involved in the aggressive phenotype of glioblastoma, and to guide new treatment strategies.ResultsWe performed a whole-genome integrative analysis of methylation and gene expression profiles in 40 newly diagnosed glioblastoma patients. We also screened for associations between the level of methylation of CpG sites and overall survival in a cohort of 50 patients uniformly treated by surgery, radiotherapy and chemotherapy with concomitant and adjuvant temozolomide (STUPP protocol). The methylation analysis identified 616 CpG sites differentially methylated between glioblastoma and control brain, a quarter of which was differentially expressed in a concordant way. Thirteen of the genes with concordant CpG sites displayed an inverse correlation between promoter methylation and expression level in glioblastomas: B3GNT5, FABP7, ZNF217, BST2, OAS1, SLC13A5, GSTM5, ME1, UBXD3, TSPYL5, FAAH, C7orf13, and C3orf14. Survival analysis identified six CpG sites associated with overall survival. SOX10 promoter methylation status (two CpG sites) stratified patients similarly to MGMT status, but with a higher Area Under the Curve (0.78 vs. 0.71, p-value < 5e-04). The methylation status of the FNDC3B, TBX3, DGKI, and FSD1 promoters identified patients with MGMT-methylated tumors that did not respond to STUPP treatment (p-value < 1e-04).ConclusionsThis study provides the first genome-wide integrative analysis of DNA methylation and gene expression profiles obtained from the same GBM cohort. We also present a methylome-based survival analysis for one of the largest uniformly treated GBM cohort ever studied, for more than 27,000 CpG sites. We have identified genes whose expression may be tightly regulated by epigenetic mechanisms and markers that may guide treatment decisions.

Background: Genomic analysis will greatly benefit from considering in a global way various sources of molecular data with the related biological knowledge. It is thus of great importance to provide useful integrative approaches dedicated to ease the interpretation of microarray data.

The Gene Ontology (GO) is a controlled vocabulary widely used for the annotation of gene products. GO is organized in three hierarchies for molecular functions, cellular components, and biological processes but no relations are provided among terms across hierarchies. The objective of this study is to investigate three non-lexical approaches to identifying such associative relations in GO and compare them among themselves and to lexical approaches. The three approaches are: computing similarity in a vector space model, statistical analysis of co-occurrence of GO terms in annotation databases, and association rule mining. Five annotation databases (FlyBase, the Human subset of GOA, MGI, SGD, and WormBase) are used in this study. A total of 7,665 associations were identified by at least one of the three non-lexical approaches. Of these, 12% were identified by more than one approach. While there are almost 6,000 lexical relations among GO terms, only 203 associations were identified by both non-lexical and lexical approaches. The associations identified in this study could serve as the starting point for adding associative relations across hierarchies to GO, but would require manual curation. The application to quality assurance of annotation databases is also discussed.

Type III epithelial–mesenchymal transition (EMT) has been previously associated with increased cell migration, invasion, metastasis, and therefore cancer aggressiveness. This reversible process is associated with an important gene expression reprogramming mainly due to epigenetic plasticity. Nevertheless, most of the studies describing the central role of epigenetic modifications during EMT were performed in a single-cell model and using only one mode of EMT induction. In our study, we studied the overall modulations of gene expression and epigenetic modifications in four different EMT-induced cell models issued from different tissues and using different inducers of EMT. Pangenomic analysis (transcriptome and ChIP–sequencing) validated our hypothesis that gene expression reprogramming during EMT is largely regulated by epigenetic modifications of a wide range of genes. Indeed, our results confirmed that each EMT model is unique and can be associated with a specific transcriptome profile and epigenetic program. However, we could select some genes or pathways that are similarly regulated in the different models and that could therefore be used as a common signature of all EMT models and become new biomarkers of the EMT phenotype. As an example, we can cite the regulation of gene-coding proteins involved in the degradation of the extracellular matrix (ECM), which are highly induced in all EMT models. Based on our investigations and results, we identified ADAM19 as a new biomarker of in vitro and in vivo EMT and we validated this biological new marker in a cohort of non-small lung carcinomas.