Hi,

I would not assume an efficiency of 100% without any data. Another program that can calculate sample-specific efficiencies is LinReg and I think further programs are available. I can recommend the webpage: http://www.gene-quantification.info/

Nevertheless you may be right that TaqMan-based assays show a low increase in fluorescence thus causing problems in programs calculating the efficiency from the slope of the amplification curve. In this case I would use serial dilutions at least if I would have a sample with "high" expression of the target genes. If you dilute this in 4-5 steps and include the samples in one run of each gene you do not have that much additional work :-)

Good luck

Estimating the amplification efficiency from the log-linear part of the amplification curves is very error-prone. Errors in estimation of the efficiency has a huge impact on the result. Consider follwowíng example: E = 1.8 (0.1) and ct=20 (0.1) (values in brackets are the standard errors), then the standard error of E^ct is 142000, where E^ct is about 127000, so this is a coefficient of variation of more than 100% !!

Take-home message: if you have uncertainties in E, you are in big trouble. The only remedy is to make as sure as possible that E (at least in the early cycles, up to the ct-value!) is as close as possible to 2.0. If you have reasons to assume that E

Dear Norman and Jochen, thank you so much for your useful advice and explanations! Up to now we didn't find any software that can calculate the PCR efficiency derived from amplification curves obtained with TaqMan probes. According to the manual of the LinReg software it only work for non-cumulative fluorescent detection techniques such as SYBR green. Therefore, we will now determine the efficiency derived from a standard curve. 

Thank you,

Kindest regards,

Audrey

This I do not understand: "According to the manual of the LinReg software it only work for non-cumulative fluorescent detection techniques such as SYBR green."

Why? I mean, the change in fluorescence between adjacent cycles is still proportional to the efficiency, and I do not see why this shouldn't be used to estimate E. Furthermore, at least the SDS exports the amplification curves as "delta Rn"-values what actually is already this change.

In the disclaimer of the manual I read "The procedures in this program are to be used for fluorescence data resulting from monitoring the PCR reaction with an intercalating dye like SYBR Green or a related chemistry. They are not valid for hybridisation probes that become fluorescent during the elongation by polymerase and of which the fluorescence accumulates." - I have two problems with this information: (1) They do not mention hydrolysis probes (not even a related chemistry) and (2) just the mentioned hybridisation probes do not show an "accumulation signal" but just as SYBR Green a signal that is always proportional to the amount of the amplicon in the reaction.

So you might not be able to answer this (but you might think about it), but maybe one of the LingReg authors will read this and help you out.

Hi Jochen, we just found an interesting article (Tuomi et al., 2010; Bias in the Cq value observed with hydrolysis probe based quantitative PCR can be corrected with the estimated PCR efficiency value) describing well the non-cumulative and cumulative nature of fluorescent detection technologies and TaqMan probes belong to the class of cumulative fluorescent detection technologies. The Taqman probe binds to DNA and is digested by the polymerase and then becomes fluorescent and stays like this, the fluorescence you see is an accumulation of the probe that has ever bound to the DNA. A bit further in the article they write the following for PCR efficiency derived from amplification curves:

The estimation of the PCR efficiency from the slope of the exponential

phase of the amplification curve, plotted on a log (fluorescence)

scale is directly derived from the logarithmic transformation of the basic kinetic equation of PCR (Eq. (1)) [34]. The application of this method seems, therefore, to be limited to non-cumulative fluorescence data [24]. However, the simulation of the estimation of the PCR efficiency from cumulative fluorescence data shows that the estimated PCR efficiencies are only positively biased when they are derived from subsets with low cycle numbers (Fig. 2C). Even with a low PCR efficiency of 1.4, already after 14 cycles, the estimation error is below

1%. With maximum PCR efficiency, this number of cycles is sufficient

to reduce the bias to 0.1%.

On the website of TaqMan they explain how to determine the efficiency with standard curves. I asked them the same question as I did here whether there is software that can be used to determine the efficiency from amplification curves obtained with Taqman. If not, we will use the method they suggest on their website just to be sure.

Thank you for the reference. I still doubt the usability of this study. Firstly they do not even try to resolve the "problem" of cumulative signal generation by taking the signal increase per cycle. Secondly, they observe a systematic difference between SYBR Green and hydrolysis probes, what is hard to explain (*). I strongly belief that these differences are due to unspecific trends in the signal, possibly caused by changes in the buffer conditions, excimer formation and whatelse. Thus, a careful detranding of the curves is essential to get accurate estimates for the efficiencies - and this is all not mentioned or discussed in the paper.

(*) Surely the steric hinderance may reduce the efficiency, but this has - to my knowledge - never been shown. On the other hand, the negative effect of SYBR Green on the efficiency has been shown. At best it's a stalemate situation.