2 replies to this topic

[Samantha]

hello

I want to know how to define whether a sequence contains a CpG island or not..I mean the critical parameters to define a CpG island. Are there any common rules?

[pcrman]

Hi Samantha,

The difinition of CpG island is evolving. Generally it just refers to some areas in the genome that are CpG rich. Check this post for discussion on this topic

http://www.protocol-online.org/forums/inde...?showtopic=3160

Usually, we use some kind of CpG island prediction program to find out if a input seqeunce contains any CpG island. For such program please click here

[methylnick]

pcrman, on Mar 11 2005, 10:18 AM, said:

The difinition of CpG island is evolving.  Generally it just refers to some areas in the genome that are CpG rich.  Check this post for discussion on this topic

Hi Samantha,

I second pcrman's response and stress that the definition has evolved to a more stringent criteria, that of Takai and Jones' definition. To reiterate what was in the link Pcrman referred to, a CpG island is now defined as being a length of DNA greater than 500bp, GC content greater than 55% and Obs/Exp CG ratio of greater than 0.65.

This criteria will exclude all repetitive elements that are thought to be methylated at ruled out as having any biological function (I would beg to differ but the reviewers of my paper think otherwise ).

The criteria now picks functional CpG islands, ones that are more likely to be associated with gene regulation.

Hope this adds to Pcrman's reply.

Cheers

Nick