I am measuring the levels of gene x after transfection of cells with a plasmid and I want to compare them to the levels in mock-transfected cells. I did three different transfection experiments, in different days (three biological replicates). I then quantify expression through RT-qPCR.
How can I combine the data from the three biological replicates? I was applying the ∆∆Ct method to each biological replicate separately, and then combining and analyzing the final fold change of the three replicates in a one-way anova, with turkey's test for significance calculation.
My supervisor suggested a different strategy: calculate the ∆Ct values (Ct mutant - Ct housekeeping) of the three biological replicates, then calculate the difference (∆∆Ct) of each ∆Ct with the average ∆Ct of the mock-transfected cells (∆Ct(transfected cells; replicate nr. x) - ∆Ct(mock; average of replicates)). Then, calculate the 2^(-∆∆Ct) of each ∆∆Ct obtained.
In both ways, at the end three fold change values are obtained (one for each biological replicate). However, the variance is much greater in the second method and, for example, there's no significative difference (one-way anova, turkey's test), while the significance is reached with the first method.
My question is: which method is the best? And, possibly, why?
p.s. I really hope I made myself clear, it's not easy to explain with words what I mean. Please tell me if you cannot understend some parts of the question.