We measured the induction of a series of genes in purified subsets of T cells cells, by polyclonal stimulation through the TCR. Since the conditions varied slightly between experiments, we calculated the fold difference between subsets in each experiment by using the highest expression attained in each experiment as the reference to normalize results. The subset with the highest expression in each experiment had therefore a DeltaDelta Ct=1, and the other subsets had a DeltaDelta Ct that was a percentage of 1. Our results were very consistent throughout the different experiments. Has this been done before? Has any justification of this been published?
Since the issue of calculating the delta-delta Ct is simply math, if your math is correct, I don't see the problem with your choice of reference; however, I think the Pfaffl method (2001 paper, I believe) discussed in the MIQE guidelines is a more accurate way of quantifying the change. The Pfaffl equation is likely the way to go, since that's accepted in the field and recommended by qPCR's inventor.
To directly answer your question, no, I haven't seen it before, and no, I haven't seen any justification for doing the math the way you have.
How about using the basal expression instead of highest expression? Did you try using basal expression as reference value and calculate delta delta CT? I think basal level of genes under unstimulated conditions will be better control value to calculate delta delta CT for induced genes.
Thanks John and Sandeep.
Sandeep, our issue is that the basal expression in many cases is inexistent. What do you do in such cases? Can we just arbitrarily assign a zero to the delta Ct?
In that case take ct value 35 and calculate the delta ct and delta delta ct and calculate fold change. As your target is undetectable so taking cycle number 35 almost indicates undetectable expression.
- Badges
- Science topic