Amplification dwindles using little template RT-PCR - (Apr/08/2009 )

Is it typical that when you use a low amount of starting template (<300pg) that your Ct values start to vary more. My higher dilutions' Ct values are nice and tight, but the less template I use the higher the std. deviation gets. Is this normal?

Also, what is an acceptable std deviation if doing triplicate of each sample?

Ahrenhase on Apr 8 2009, 03:18 PM said:

This is typical for us, especially when Ct values are above 35. With low template, our samples may range from a Ct of 35 to 38, while increased template samples are consistent at lower Ct values with differences in Ct values that are less than 0.2. I don't think that accurate estimations of the linear phase can be made when the Ct values are so high.
As for standard deviation, this depends on the Ct differences between your samples and the significance you are trying to achieve. The standard deviation should be

Well, I was just diluting them down to test primer efficiencies. Since the first 3 dilutions came out very nice, and the 1x would be the dilution I'd use when testing relative expression, I will assume that its a template issue and not an ineffecient primer. All my peaks on my melt/dissociation curve came out nice too so that leads me to believe it's the template concentration.

I just don't understand why the instruction manual my RT-PCR machine comes with tells you to run efficiencies on the primers by diluting the template down 6 times and creating a linear graph of the change in Ct. If you ran just one template dilution in triplicate and then did a melt curve, wouldn't that right there tell you whether your primers were working or not?

Ahrenhase on Apr 9 2009, 12:30 PM said:

Calculation of the primer efficiency is not just to tell whether or not the primer is working, but to tell how well the primer is working. Primers can work, yield single peak melt curves, but have poor amplification efficiency. This is a problem if your reference gene primer sets have different efficiencies in that you are getting a false representation of your target level if the efficiency is low. See some of the literature by M. Pffafl.