DNA methylation leading gene silencing is found in many types of cancers. COBRA (combined bisulfite restriction analysis) is one of the most common methylation quantification methods employed by regular laboratories. However, the traditional COBRA is focused on a few restriction enzymes, such as BstUI and Taq^αI. Here, we present Methyl-typing service,viaa free web server (http://bio.kuas.edu.tw/methyl-typing) designed to provide the restriction enzyme mining data for the methyl-cytosine-containing sequences following bisulfite-conversion. Gene names, accession numbers, sequences, PCR primers, and file upload are accessible for input. For gene name input, the promoter sequences are retrieved from DBTSS (database of transcription start sites). All the inputs are able to perform CpG island search and visualize in an interactive manner. Restriction enzymes downloaded from REBASE provide a comprehensive list of enzymes for CpG- and GpC-containing recognition sites. Moreover, the insulators CCCTC-binding factor (CTCF), neutralized by methylation-binding site database (CTCFBSDB) is implemented. Methyl-typing also provides bisulfite T stretch function to predict possible poly T in computer-generated bisulfite-converted sequences. Six representative enzymes were successfully tested by COBRA on oral cancer and control studies. To our knowledge, this is the first application enabling comprehensive COBRA-restriction enzymes mining, and serving as a visualization tool.

Traditional COBRA vs. Computation

Demonstration of methyl-typing by six PCR-RFLP tests