Statistical analysis of DNA methylation patterns of tumor suppressor genes in breast cancer

The purpose of this study was to conduct the first-ever comprehensive analysis of methylation patterns of tumor suppressor genes (TSGs) in breast cancer. The authors first identified differentially methylated (DM) sites between tumors and matched normal tissues in both Alive and Dead samples. They then analyzed co-methylation patterns related to these DM sites and reported corresponding TSGs and non-TSGs. First, tumors had more heterogeneous methylation sites than normal tissues (40% vs < 10%) in both Alive and Dead samples. Second, there were significantly more DM sites in Dead than in Alive samples. Third, as for co-methylation patterns, in normal tissues, some DM sites tended to have strong co-methylations with many CG sites. In tumor tissues, some of these strong co-methylations were lost, and some new co-methylation relationships were developed. These patterns were seen in both Alive and Dead data. Fourth, there were more co-methylation changes between normal and tumor tissues in Dead than in Alive samples. Thirty TSGs and 92 non-TSGs were identified as having notable differences between Alive and Dead data. Finally, seven TSGs were involved in many co-methylation changes between normal and tumor tissues. These seven TSGs were hub genes in different networks.