I am trying to do a qPCR analysis of a miRNA in the rat brain that needs to control for treatment, sex, and siblings and I'm having trouble w/stats. Usually I do the delta delta ct method, but having the within-litter comparison, plus treatment, plus controlling for the endogenous control is making this challenging.
Why not simply normalise every sample, then compare them directly in whatever combinations you choose?
ddCt methods seem to rely inordinately heavily on appropriate pairings of samples, and it is not always clear what those pairings should be (or even whether such pairings should be made at all).
dCt values (i.e. the gene of interest for a single sample, normalized to that single sample's reference gene(s)) are sufficient for comparisons.
So: dCt for all your samples, then compare the dCts of one litter with those of another and see if there's a significant difference.
Thanks for your help! I was really struggling using the ddCT method thinking that it was the only acceptable way to plot. It really was difficult to pick that appropriate pairing of samples.
Honestly, I prefer to use relative quantities (RQs): basically convert all those confusing log values to linear values, so that I can see at a glance exactly how my numbers stack up, and use entirely conventional statistical comparisons etc. It doesn't change the interpretation, it just makes it easier to avoid making mistakes.
And I'm a simply creature who fears logarithms, if I'm honest.
RQs are essentially "efficiency(lowest Cq - sample Cq)", but I've included an excel file (using made up numbers) showing a worked example via dCt and RQ.
Standard curve: make a serial dilution of your cDNA (or of a purified PCR product) and measure the gradient of Cq values when you use those in qPCR.
If you plot "log(dilution factor)" against "Cq value" (X and Y respectively) you should get a slope of around -3.32.
The efficiency is 10(-1/gradient) , i.e. for -3.32, 10(-1/-3.32) = 2.00
Efficiency is effectively "signal increase per cycle", so a 100% efficient PCR should show a neat doubling with each cycle (efficiency 2.00). Am 85% efficient PCR will only show a 1.85x increase in signal each cycle (efficiency 1.85).
Ideally all your efficiencies will be 95-100% (1.95-2.00), but it never hurts to check.
Excellent, thank you! I appreciate you taking the time to explain it all. I hope to pay it forward someday :)
- Badges
- Science topic
- Similar topics
- Biological Science
- Biochemistry
- Biomolecules
- Nucleic Acids
- RNA
- Small RNA
- microRNA