A genome-wide study on differential methylation in different cancers using TCGA database

Background: DNA methylation is the main epigenetic mechanism driving changes in phenotype without altering genotype. Since the end of the seventies the role of methylation in cancer has become increasingly clear. Objective: The aim of this work is to shed light on the impact of methylation events on cancer cells, providing evidence that differential methylation in small regions, mostly characterized by hypermethylation, affects gene regulation while differential methylation in large genomic regions, mostly characterized by hypomethylation, affects chromosomal organization. Methods: By exploiting a solid computational and statistical analysis, methylation maps of cancer and normal samples in six different cancer types were studied, looking for those genomic regions showing differentially methylated patterns between the two conditions. Results: Through a chromosome sliding windows approach, a set of differentially methylated genomic micro regions of size 2 K bp and macro regions of size 1 M bp, were identified. Micro regions are mostly linked to functional elements while macro regions are mostly linked to nuclear chromosome organization. Results discussed in previous works were confirmed, providing clear evidence that hypermethylation mainly occurs in significant micro regions while hypomethylation mainly occurs in significant macro regions. Interestingly the presence of differentially methylated regions common for six different cancers were identified and some unexpected and previously unexplored peculiar methylation patterns were also found. Conclusions: The effective and robust computational and statistical methodology presented in this work can be used to shed light on the role that DNA methylation plays in cancer and in other non malignant diseases and can be customized to study differentially methylated patterns in specific areas of interest of the genome both at a small scale and at a large scale.