real-time PCR products on gel look good but melting curve is horribel - (May/25/2007 )

Dear all,

I try to "reanimate" an iCycler machine which has not been used for more than a year. I run the first experiment with a GAPDH primer pair and the melting curve looked not so bad as only in the NTC could be seen an extra peak. I run the experiment again with 5 different primer concentrations but this time the melting curve looked rather ugly for all the samples tested (including NTCs). Than I repeated exactly the same experiment as at the first time and the melting curve was even horrible. I thought that maybe the problem is with my primers so I got another primer pair from a colleague. I did the melting curve and it showed many peaks again. Than I run the products and for my biggest surprise there was only one very specific band on the gel, no unspecific products or dimers at all. Do you have any suggestions what could be the problem? Could the PCR-mix cause this or the PCR machine, or the problem is with me?
I have no huge experience working with iCycler so I would highly appreciate any help,
Grumpy

There are many possible solutions. We had a lot of problems with our I-Cycler once, as the block did not heat continuosly any more and the ramping rates were really funny - this could lead to melting curves like yours but doesn't necessarily have to affect the product. I'd have a BioRad technician check it...

Dear krümelmonster,

Thank you for your help. In the meantime, I had a possibility to run the same experiment simultaneously on our iCycler and on a MyCycler in an other Department and as the result was okay on this other machine but still awful on our iCycler, I think that I can exclude other factors but the machine itself...
Thank you again!