Pyrosequencing Bisulphite Conversion Control - How exactly does this work? (Mar/15/2008 )
The more and more I have thought about the pyrosequencing bisulphite conversion control the more confused I have become. I am hoping that somebody will be able to save me from my confusion!
I understand the concept of the control i.e. If you have not had full bisulphite conversion C nucleotides that are not part of CpG sites should have been fully converted to T. The idea of the control is that if incomplete conversion has occured a and C is represented in your sequence a C peak will be present in your pyrogram at the C dispensation for the Bisulphite conversion control
My issues is that how does this control actually work as the primers that you design for pyrosequencing as specfic to converted DNA and therefore there technically be no amplification of genomic seqeuence. If there is no amplification of genomic sequence then how is it posible to obatin a pyrogram with a C present? (apart from at CpG sites)
Like I said I have proabably just got really confused but would be greatful for any help. 
Thanks
The control issue You're talking about deals with the bisulphite treatment itself.
Briefly, bisulphite, at specific concentration, determines hydrolitic deamination of cytosines which are not methylated, giving uracil.
Uracil will match with "A" of your specific bisulphite primer during the first cylce of PCR.
During the second cycle the A-containing template elongated from the converted genomic DNA will match with the other primer eventually giving the C ---> T conversion. The final result is that DNA looses a "base of the code" giving a highly redundant sequence.
Here we come at Your issue.
Let's assume that the efficiency of the conversion is less than 100% (i.e. 95%). Since the conversion process is dependent on many factors (i.e. the "sequence" itself and the "denaturation"), not all the unmethylated Cs would be affected by the bisulphite conversion. Thus, few unmethylated Cs of the amplicon could "remain" Cs.
Also, sometimes, depending on the primer sequences, a specific primer could still amplify a not-fully-converted DNA whitout results in untreated genomic DNA as control even if they have "few" (one, two...) mismatches.
Of course this issue can invalidate the methylation analysis as you can't be "sure" that Cs are "really methylated Cs".
Now, the pyrosequencing protocols include on purpose the C-T control (sometimes more than one, depending on the length of the amplicon) to verify the proper conversion of the DNA and of course this control involves Cs which are not incuded in CpGs dinucleotides.
I hope that could help.