Standard curve and interplate calibrator

Here we discuss the use and processing of inter-plate calibrators

Moderator: MultiD Support

Standard curve and interplate calibrator

Postby Pasajee » Thu Jul 23, 2009 4:06 am


I already runned all of my plates, but I have a problem analyzing them. I'm interested to use GenEx to analyze them. So, I have some questions before I start downloading GenEx to try.

I have 96 samples (32 samples, but I have 3 biological replicates).
I runned 3 technical replicates per plate therefore I have to run one plate for one gene. (in 384 well-plate)
For each plate, I have a standard curve for that gene (5 dilution of my sample cDNA pool, for each dilution, I runned triplicates) and 96 samples in triplicates.
So, I can calculate relative quantity of each gene from the standard curve.
I used the calculated relative quantity of my 96 samples for 8 reference genes (from 8 separeated plates) to input in geNorm program to select two reference genes for normalization.
Then, in my GOI plate, I have 3 standard curves which are standard curve of my GOI, standard curve of RG1 and standard curve of RG2.
I also have two samples which were runned by my GOI, RG1, and RG2 on the same plate. But, for the rest space, I runned 96 samples in triplicates with only my GOI.
I would like to analyze my data by dividing relative quantity of my GOI by the geometric mean of relative quantity of RG1 and RG2. But, my concern is that my GOI, RG1 and RG2 were all runned on the different plates. And from the standard curve of RG1 and RG2 on my GOI plate, I can say that the amplification efficiency of RG1 and RG2 are quite not exactly the same as in the RG1 and RG2 plates.

My question is if I use GenEx to analyze my data, can I use my standard curve or each dilution in my standard curve or two samples that I runned with all genes on the same plate as the interplate calibrators?

Thank you in advance,

Posts: 1
Joined: Thu Jul 23, 2009 2:33 am

Re: Standard curve and interplate calibrator

Postby Mikael Kubista » Thu Jul 23, 2009 12:12 pm

Dear Pasajee,
You have obviously performed a rather complex study, though, really not unusually complex, but rather in line with how modern studies are performed. There is no way you can analyze your data with a simple formula like CT. Instead you must use the approach of GenEx, where you perform the necessary corrections sequentially (see: G.I.T. Laboratory Journal (2007), 9-10, 33-35, available on or GenEx manual).

Importantly, you must adjust the data from the different plates to the same level by inter-plate calibration. Interplate calibration is available in GenEx 4.4.8 and even more options are available in GenEx 5.0, which will soon be released (Beta versions are already being distributed).

You must treat the technical and biological replicates differently, which is done by GenEx

You extract PCR efficiencies from the standard curves to correct the measured values. It is important to estimate the uncertainty of the calculated PCR efficiencies, which is done by GenEx. I suspect the variation you see in PCR efficiencies estimated in different plates is not significant. Better strategy that running separate standard curves on each plate is to make a single thorough standard curve from which a single PCR efficiency is estimated for all corrections.

If you analyze multiple RG’s you should consider analyzing the data with multivariate methods such as clustering, PCA and SOM, available in GenEx.

Good luck!
Mikael Kubista
Posts: 455
Joined: Tue Jul 01, 2008 12:28 pm

Return to Interplate calibration

Who is online

Users browsing this forum: No registered users and 1 guest


MultiD Analyses

Home of the GenEx analysis software