I am doing PCR experiments with different target genes and different potential reference genes (housekeeping genes). I have read that to apply Delta Delta Ct method, the Ct value of reference genes and target genes shoud be close. Any suggestion about this?
It's a physics 'fulcrum' issue (of sorts) - but I'm actually not a big believer in the fulcrum paranoia on this... more important are good efficiencies and establishing valid dynamic ranges of measurement for all targets and reference genes.
Some reference genes are hyper-abundant (16S, 18S)... so their standards and samples need to be diluted more to get within the same Cq range as target genes of interest. But... as long as good measurements have been made for both target and reference within their respective valid ranges, the likeness of their Cq values shouldn't matter. Try to use the largest dynamic range possible for all standard curves. Sometimes you're lucky if you can get 2 logs. Other times, 5 or more logs is easy (i.e. with abundant target or ref genes, plasmids etc.).
Generally, it seems that Cq values between 14 and 35 are the most "reliable" (which constitutes ~6.32 log range for the real range of usefulness of qPCR). But, know your efficiency of amplification for all targets and reference genes, always.
Jack, thanks a lot for your information and your answer. I agree with your comment, the most important thing is similar efficiencies, the consitency of Ct values and reliable Ct values (from 14 to 35).
Jose, your summation seems dead on. Additionally, to really assess efficacy of a candidate housekeeping gene as target, it's important to:
1) Prepare 10-fold dilution curves of (at least) one DNA sample which is representative of the DNA isolation/prep conditions you'll be using in the final assay (being sure that you do a broad enough range to ensure you can get overlap of at least three dilution points between the below-described experiments:
a) Run qPCR on triplicates of this sampledilution curve, using primers specific for your target of interest;
b) In parallel, run qPCR on this curve (triplicates again) using primers specific for the candidate housekeeper gene;
2) Calculate [(Target Gene Mean Ct) - (Housekeeper Gene Mean Ct)] for each dilution step (or vice versa, of course);
3) Plot these values on a Ct vs Dilution graph, and perform a linear regression;
4) Also, plot the results of your housekeeper gene on a Ct vs. log[concentration] graph;
A suitable housekeeping gene should present all of these:
--qPCR results plotted in (#4) should be linear in the range of concentrations representing what you reasonably expect to see from testing your target gene.
--The plot of ddCt (calculated in #2 and plotted in #3) should yield a linear regression with a slope no less/greater than -0.1 to +0.1.
--Anything other than these two results will produce bias error in your normalizations of target gene results.
Thanks a lot for your answer and comments, Larry. All the information was very usefull.
In addition to Larry's excellent comments above -- in order to see the entire behavior of the dynamic dilution range(s), serial 1:2 dilutions (for about 25 points) is the most informative way to view your standard curves the first time. Once, done, then hone in and use the "sweet" ranges.
Hopefully, when you do this, you will also find that "The plot of ddCt (calculated in #2 and plotted in #3) should yield a linear regression {[difference in slopes]} no less/greater than -0.1 to +0.1" that Larry speaks of. It is an age-old ABI User Bulletin #2 edict of sorts.
Thanks for all your comments. However, if the ct of your reference gene is 12 and the ct of your target gene is 24 could you go on with your expression analysis ?
- Badges
- Science topic