Yes you can but you should also compare the expression of your gene of interest against an endogenous gene for both cases (before and after treatment). You should have an endogenous gene because you need to verify how your treatment afects gene expression against a standard.

You can only use 2^-ddCT if your primers efficiency are the same, if not you should pfaffl method.

I use 2^-CT for comparisons between genotypes and treatments in mouse and mouse cell biology. It seems to work, and there are methods to correct for differences in primer efficiency when using 2^-CT, and when normalizing to a control gene that appear functionally equivalent to using an efficiency corrected DeltaDeltaCT calculation. That said, differences in efficiency generally only slightly affect the magnitude of an observed change (in my experience).

I prefer to analyze RNAs and cDNAs processed in parallel to avoid issues with possible differences in RNA integrity/quality or the presence of RT or PCR inhibitors between purification or cDNA syntheses. If you normalize to the loading contol you should be able to get away with a between groups comparison (e.g. via ANOVA).

Be on the look out for signs of obvious inexplicable differences between the sample sets, though (e.g. the loading control varies by sample group).

Dear Methee,

As mentioned by Andre, as long as you run a housekeeping gene during your PCR and that the amplification efficiencies of the primers for the HKG and your gene of interest are similar and close to 100 %, you can use the 2^-ddCt method.

Another requirement if you want to compare the expression before and after treatment is to run the samples for both the HKG and your gene of interest on the same PCR plate, so that the PCR conditions are exactly the same.

Best,

Damien

A simple approach is to pick one sample without drug treatment as the reference sample (Rq =1).  Then calculate the Rq for every sample.  Then determine the avg and SD for all samples in the two groups (before and after treatment).  This would allow you to compare the groups by ANOVA or T-test.  However, if you just calculate the change in gene expression after drug treatment for each patient individually, you will not have two groups and will have to use a different statistical test.  This might be better if you notice a consistent change in gene expression, but the starting values are very different between patients. I haven't done calculations this way, but maybe someone else will know the best test to use.

Michael

I would like to ask you one thing. If Mathee picks one sample as reference he would have to assume that all his RNAs are equal and the RT efficiencies for each sample are the same. Don't you think he would have to assess a HKG to normalize any inconsistency?

A housekeeping gene is definitely needed for all samples.  Delta Ct is gene of interest minus HKG for each sample. Then delta delta Ct compares the change in this quantity between samples.  Rq would be calculated either by assuming an amplification efficiency of 2 or using a standard curve to be more precise.  I think the issue is whether to determine delta delta relative to a single sample in the entire set, or whether to determine delta delta within each patient before and after drug treatment.

Dr. Brian Bower

According to your reply, that means I can use 2^-dCT to compare the different gene expression between pre and post treatment or compare different groups such as group A and B during pre-treatment. 

Yes. I don't see any problem with that. Especially if you're using a proper loading control.