Hi Jernej - 

Sorry but I am a bit confused. I assume you are looking at the reverse strand, in that G's on the trace represent methylated C's in the original sequence. My question is why is the G signal so strong in your second plot? If those peaks were of equal height as the other bases, would the two plots look more similar? It has been a while since I have performed Sanger seq. 

Best,

Brian

Dear Brian, sorry for a confusion. It obviously wasn't clear enough, but yes, the sequences were made by reverse primers. And high Gs are actually Cs, thats why my question about non-specific C signals.

Dear Jernej, Brian, and Blanca,

Thanks for mentioning our paper on over- and under-conversion, Blanca!

To this conversation, I'll just add that our experience has been that high temp/short duration conversion is key to getting reliable bisulfite data. This is almost certainly because high temps (80C, in our hands) help to keep the DNA in a denatured state, such that the probability of conversion for unmethylated cytosines is more nearly uniform across sites.

By contrast, under protocols with lower or fluctuating temperatures, under which stochastic variation across sites in the total duration of single-strandedness can yield within and among molecule variation in the efficiency of conversion. High temperatures also allow for higher concentrations of bisulfite to go into solution --- and to stay in solution over the conversion interval.

Lately, we have had good success with applying the principles of high temperature/high molarity/short duration conversion with the SIGMA Imprint kit. That said, we have also had good success with "homemade" solutions that achieve high concentration of the bisulfite ion.

I'd be happy to chat more about making effective conversion solutions in one's own lab, should that be of use!

Best wishes with your experiments, 

Diane Genereux