Mona,

I think you've got the analysis exactly right! The reason it can't be absolute, is because you don't know exactly how many copies of mRNA are in your sample, or copies for each gene. I know there are ways to do this, using a reference sample with a known amount of RNA. However many people just want to look at changes between samples, which is why relative changes and arbitrary copy numbers are good enough, and aboslute numbers aren't usually necessary.

I think this paper explains RTqPCR analysis nicely! Good luck!

Schmittgen, T. D., & Livak, K. J. (2008). Analyzing real-time PCR data by the comparative CT method. Nature Protocols, 3(6), 1101–1108. http://doi.org/10.1038/nprot.2008.73

Sara

Some cases assessing "copy numbers" are informative, i.e. that is how HIV can be screened for drug response, or BCR-ABL for minimal residual disease. If you're doing expression analysis, it's not that informative to find x copies, you can get a similar answer using relative expression without the use of extra standard(s). If you are using relative expression, you just need to find one (or more) stable gene to normalize expression and you can measure gene expression changes. But with copy numbers, you would need to constantly add standard curves for each gene hence being more costly. If you're going to use the reference gene approach, make sure that you have experimentally verified this as people make the assumption that commonly used reference genes are stable in their condition which i did not find to be true in my doctorate studies.

Majority of the people are only interested in knowing before and after treatment comparison, thus they do relative. In some cases, you would like to know the actual numbers. So in short:

You actually can. Accugen is pioneering this technology, www.accugen.com.au

Have a look at their paper

https://www.ncbi.nlm.nih.gov/pubmed/26956612