Methylation type introduction

All methylation data in MethyCancer were integrated from 5 different sources, namely HEP, MethDB, UHN, Columbia University and BIG, and were classified into 7 methylation types according to the properties and sources of the methylation data.

1. Methylation region:
The genome-scale dataset of methylated domains and unmethylated domains in human brain tissues from Columbia University were classified as 'methylation region'.

2. Methylation content / Methylation pattern / Methylation profile:
Methylation content represents the total amount of methylated cytosine in the genome, expressed either as a fraction of cytosines, or as a fraction of CpG dinucleotides (Laird, 2003).

Methylation pattern refers to a series of CpG dinucleotides located in cis on a single DNA strand. DNA methylation patterns can be determined by bisulphite genomic sequencing of subclones. Methylation-specific polymerase chain reaction measures the presence of one particular methylation pattern in a pool of DNA.

Methylation profile refers to the measurement of DNA methylation at multiple sites throughout the genome.

Data from MethDB were classified into methylation content, pattern, and profile, while data from HEP were grouped into methylation profile.

3. CpG Island I, II and III:
The CpG island (CGI) clone sequences from BIG and UHN were mapped onto human genome (NCBI Build36) using BLAT and BLAST. All clones were clustered into genomic loci (Loci). To verify the experimental CGI sequences and identify novel candidate CGIs, CGI predictions on the whole human genome were performed using CpGi130. If a Loci matches with CGI prediction, we define it as CGI type 1 (CpG Island I). If not, we predicted CGI on its representative clone. If the representative clone of a Loci matches with CGI prediction, we define it as CGI type 2 (CpG Island II). The rest are defined as CGI type 0 (CpG Island III). For more details, please look at the document of CGI analysis.

Laird P. W. (2003). The power and the promise of DNA methylation markers. Nat Rev Cancer 3: 253-66.