The centrosome plays a central role for cellular signaling and is critical for several fundamental cellular processes in human cells. Centrosome abnormalities have been linked to multiple solid tumors and hematological malignancies. We sought to explore the potential role of the DNA methylation, a critical epigenetic modification, of centrosome-related genes in different cancers. The 450K array DNA methylation data and RNA-seq data were downloaded for ~4000 tumor samples and ~500 normal controls from The Cancer Genome Atlas (TCGA) project, covering 11 major cancer types. Cancers with more than 30 normal controls were retained for analysis. Differentially modified CpGs of centrosome genes were identified, and cancer-specific epigenetic models were developed using a machine-learning algorithm for each cancer type. The association between the methylation level of differential CpGs and the corresponding gene expression, as well as the co-localization of the differential CpGs and cis-regulatory elements were evaluated. In total, 2761 CpGs located on 160 centrosome genes for 6 cancers were included in the analysis. Cancer-specific models demonstrated a high accuracy in terms of the area under the receiver operating characteristic (ROC) curve (AUC > 0.9) in five cancers and showed tissue specificity. This study enhanced our understanding of the epigenetic mechanisms underlying the DNA methylation of centrosome-related genes in cancers, and showed the potential of these epigenetic modifications as novel cancer biomarkers.
Keywords: 450K array; DNA methylation; RNA-seq; TCGA; cancer detection; centrosome.