end point or real time?

4 posts / 0 new
Last post
AJ's picture
end point or real time?

I'm a graduate student about a start a project involving a lot of RT-PCR. If I have around 20 genes whose expression I have to study in 8 different conditions, which method would make more sense for me to use, considering both money and labour/time investment? Can someone give a rough idea about the relative costs, if both types of PCR machines are freely available for me to use?

SanDiablo's picture
Do you need to absolutely

Do you need to absolutely quantify the level of expression of your 20 genes or will you just be looking for +/- expression? Will you be lookin at the 20 genes from the same 8 conditions (eg a time cousre or concentration curve?) or will there be a different set of 8 conditions for each gene?

End-point PCR is easier and less expensive, but it won't give you quantitaive results, such as fold changes in expression levels. It will only tell you if the mRNA is present or not. Real time or quantitative PCR will require designing and optimizing 20 sets of primer/probes, including one or two controls for normalizing between the 8 conditions as well as between individual PCR runs. You will need to run a standard curve for each gene, including controls on every plate.

8 conditions is not bad, as you can prepare a single batch of cDNA for each condition and aliquot it out for all subseuent PCR reactions. Claculate how much you will need and make a double or triple batch if necessary. If you have to run new cDNA reactions, you will have more difficulty comparing one run to another.

On a 96 well plate, you can only run 2 different genes at once (7 stds, 1 NTC and 8 samples per each primer/probe set) because you will need to run each sample in triplicate. So 20 genes will take 20 runs, or about a week, depending on your access to the machine(s). It could take up to two-three weeks of validation, and you will likely have to repeat several of the runs for one reason or another. The data crucnhing will be the hard part, so be sure to learn all you can about normalization and use an internal control suitable to your system - preferably one whose expression level is constant over your 8 conditions. People normalize their reactions in lots of different ways, so see what your advisor believes is best.

If you don't have access to a real-time PCR machine or if your budget doesn't allow you to get 20+ primer probe sets, you can get semi-quantitative results with end-point PCR by performing your PCR reactions on a dilution series of your cDNA, then visualizing the results on a gel. This way, you can determine the fold dilution where amplification is first detectable and where amplification is saturated, which gives you a rough way to compare expression levels. You'll still need to run control reactions side by side. This approach is a lot more work, but it depends on what your objective is...merely results, or a pretty picture for publication.

Either way, if you get some notable expression changes, verufiy their levels through another method, such as a Northern, dot-blot, or in situ hybridization.

Sorry, I don't have any hard cost figures, as I worked for a rich company and never saw the bills!

AJ's picture


Pharmalance's picture
AJ wrote:Thanks!

AJ wrote:


Great Job SanDiablo
just wanted to add a little thing in case you want to use traditional PCR
you have to stop your PCR within the linear phase so that the amplification ratio will be kept (try 30 to 35 cycles)

also if you want end point PCR you may want to consider a multiplex PCR with your 8 genes + 2 housekeeping genes for a relative quantification on a sequencing machine.
check this publication it may help you

Johnson PH, Walker RP, Jones SW, Stephens K, Meurer J, Zajchowski DA, Luke MM, Eeckman F, Tan Y, Wong L, Parry G, Morgan TK Jr, McCarrick MA, Monforte J. Related Articles, Links
Free Full Text
Multiplex gene expression analysis for high-throughput drug discovery: screening and analysis of compounds affecting genes overexpressed in cancer cells.
Mol Cancer Ther. 2002 Dec;1(14):1293-304.
PMID: 12516962