In principle, you can normalize to the total RNA (=cDNA) amount, without using reference genes, meaning that I have seen plenty of publications using this normalization method and they were published just fine. [I'm not going into discussion about appropriateness of use of reference genes, it's a whole topic in and of itself; I'm just saying normalization to cDNA concentration CAN be published in high impact journals just fine]

So if you prepare all your cDNA samples and make sure they all have same concentration, you don't have to run reference genes at all.

Run NRC and NTC once, maybe on a separate plate (to save space on actual plate), and then if you're careful, you don't have to repeat them every time.

Hello Vishnutej

You are supposed to use a reference gene but it does not have to be run on the same plate as your desired gene. Your reference gene/s values will be used to normalize your data. The point is evaluation of all of the genes (desired or reference ones) must be done on the same cDNA.

You should not use pooled cDNA of all individuals for your reference gene. It is totally inappropriate. The application of pooled cDNA is to run and draw the standard curve.

Standard curve samples must be included with your plated and in every run.

For your reference gene, you need to use all individual cDNAs, same as your desired gene.

All the best

I would split the assays on plates then you have no problems because all replicates from all sames are run together and the algorithm worms fine with this setup

Hi,

1. You can load your plate using a sample maximization approach. If you can fit all your samples for the gene on one plate, then you do not need an inter-run calibrator. It is good practice to take for every gene an NTC and NRT controle. More info on sample maximization versus gene maximization can be found on the outstanding QBASE+ blog: https://blog.qbaseplus.com/four-tips-for-rt-qpcr-data-normalization-using-reference-genes

2. It is good practice to normalize every sample towards your reference gene. It makes no sense to normalize over a pool of cDNA. You want to be able to detect differences in the expression of your GOI. Hence, you will have to normalize every sample on multiple reference genes. This to not only to account for input differences (preferably, you normalize your input), but very importantly, to take into account differences in RT-activity. A prerequisite is that those reference genes are stably expressed. (More info on the same blog page).

For more information about correctly performing qPCR experiments, I can also really recommend the following paper: https://www.cell.com/trends/biotechnology/fulltext/S0167-7799(18)30342-1

Caroline Weydert . Thank you very much for the elaborated answer. I have some followup questions

In my experiment I have samples collected at 6 time points, for the same gene. For each time point I have 3 biological replicates and 3 technical replicates taken from them, for both control and treated. It makes 18 samples at each time point.

1. So do you mean, for every time point I should use cDNA of corresponding time point to run reference gene/s. ?

2. References genes should be run in triplicates ?

What Caroline said was that you could split your reactions in regard to the assays e.g. making one plate with Gene of interest 1, the next one with gene of interest 2... This approach will allow you to compare the resultsa s the conditions for all samples per assay will be virtually the same. it does not matter if there is a small diffewrence between the runs as difference will then apply again for all samples

Sven Reister : exactly!

Concerning the other question: in RT-qPCR you are comparing relative expression. So you need to assess preferably more then one reference gene per sample.

Concerning technical replicates: taking duplicates or triplicates mainly depends of how confident you are with your pipetting skills. Other sources of variation are not taken into account by making technical replicates for the qPCR experiment. The most important replicates are biological ones! What is also more important, is to think about your RT reaction. Which RT enzyme do you use, which primers, does it do what it needs to do, does it work correctly for the concentrations of RNA that you are using? More info on this can also be found on the excellent qBASE+ blog: https://blog.qbaseplus.com/is-it-better-to-pipet-duplicates-or-triplicate-reactions-in-real-time-pcr