reference gene validation

1 post / 0 new
Li Liao
Li Liao's picture
reference gene validation

Hi all,

I am designing my qrt-pcr for relative quantification using pfaffl method. First, I have to validate reference genes which are housekeeping genes (HKGs)  from literature in other bacteria (since I work on bacteria). My question is how to validate HKGs?

My idea is to compare the HKGs expression ratio in two samples (since I only have two different samples). If the HKGs (at least 3 genes) are expressed in the same ratio in two samples, that means they are most likely stable and suitable for reference in my experiment.

So to quantify the HKGs before calculating the ratio, I am trying to design absolute quantification of these HKGs using DNA standards (e.g., plasmids containing target genes) in each sample. If, e.g., I put 10ng of total cDNA from sample 1 into each reaction (containing primers for each HKG), and then calculate the copies of HKGs from their standard curves. After this, calculate the ratio using copies of the HKGs in sample 1. Likewise, do the same thing in sample 2, and get the ratio in sample 2 as well. Now I can compare the two ratios, if they are the same, or some HKGs have the same ratio, then I can choose those HKGs for reference.

It's complicated to explain. I hope I made it clear. My question is can I do the validation in this way? Is that reasonable?
If you have good idea, I am happy to try it.

Also, I know some programs like genorm and bestkeeper, which are designed to evaluate and find the "best" reference gene(s). But I am not quite clear about how they work. They seem to work only for relative data, using Ct comparison. Any thoughts?

Thank you so much!
Li