Question
Asked 14 November 2014
How I can compare a relative ratio of the basal mRNA expression of 2 genes in the same cell line by qRT-PCR?
I found in some paper they can use qRT-PCR to quantify mRNA of gene A, have x time higher than gene B expression, however I could not find how they do that in detail
I use Bio- Rad iQ5 software to calculate Tm, because we can use the Tm to determine the basal gene expression by delta-delta-Ct. However, I can change the Ct vulvae such as in these attached images. Will the change make the relative ratio between the 2 genes also change?
Pls help me explain, with a detailed mathematical formula, how to calculate the relative ratio between 2 genes in the same cell depending on the Ct value of the gene .
And how we can choose the suitable Ct value?
Thank you so much !
Most recent answer
As far as I know from many post I have read here, Jochen is an expert on this and he is right. If you want to publish this comparison data or use it as an argument for your hypothesis, you have to use absolute quantification.
But if you only want to have an idea of relative expression of a group of related genes (for example, you are working on a cytokine and there are 2 receptors for this cytokine, you want to find out which is the prominent receptor for your cell type/state) in this case you can do relative quantification. I always do this before ordering antibodies, antagonists, inhibitors etc.
Popular answers (1)
Justus Liebig University Giessen
As I said: perform an absolute quantification of the two genes using an external calibrator.
Again, just to clarify:
You can compare the relative expression of one particular mRNA between different conditions or cell types. You can do this by simply comparing the dCt values (-> ddCt). I think this is what most papers do.
A comparison between different genes (in the same condition or cell type) is not that simple. Comparing dCt values for different target genes is meaningless (well, if the differences are huge (>>10x), then a comparison will likely point in the right direction; but then only the direction is indicated, so the conclusion is qualitative, not quantitative!).
Note that the dCt value is a normalized value of a target gene. The normalizer is a (mean) ct value from one or more reference gene(s) which expression(s) serve as a loading control. A single dCt value does not tell you much. It is just a value "in outer space", without a sensible reference point. In particular, a dCt of zero does *NOT* indicate that the expression of the target gene is as high as the expression of the reference gene. It is also irrelevant or uninformative if the dCt is positive or negative. The dCt contains this unknown proportionality factor. BUT: different dCt values for the same target gene (and the same reference gene(s)) can be compared, because they contain the same (unknown) proportionality factors that cancel out in the difference of the dCt values (i.e. in the ddCt value).
If your problem is setting the threshold value:
You can place the threshold at any value where the amplification curves show an exponential increase (and where all these log-linear segments are parallel).
Shifting the threshold will shift the Ct-values. But because the curves are parallel, all Ct values will be shifted by exactly the same amount. Hence, the dCt values will not be influenced by the threshold.
As you can see from the formulas I posted previousely, there is no need to chose the same threshold for the traget and the reference gene. The proportionality factors will cancel out in the ddCt as long as the same threshold (per gene) is used for the different conditions [or the same threshold (per condition) is used for the different genes {target and reference} - what is mathematically ok but rarely sensible in practice].
3 Recommendations
All Answers (6)
When manually placing the threshold line, select a region in the linear amplification part of the curve (100-200 in your example picture). Changing the threshold line from 100 to 200 should only cause a minor difference in the result.
Ct value of Reference gene (Ct RG)
Ct value of Gene of interest1 (Ct GOI1)
Ct value of Gene of interest2 (Ct GOI2)
dCt 1 (delta Ct1) = Ct GOI1 - Ct RG
dCt 2 (delta Ct2) = Ct GOI2 - Ct RG
%RG expression for GOI1= 100 x 2^(-dCt 1)
%RG expression for GOI2= 100 x 2^(-dCt 2)
then you can calculate the relative expression ratio of these two genes.
check this out for detailed analysis:
1 Recommendation
IRCCS Multimedica
I attach you the file that I use for the calculation of relative expression, you have to insert the Ct of your genes (housekeeping gene in the first column), and then Ct of genes of interest. You need to put also the % of primer efficiency (you can have it if you do the standard curve to see the primer efficency). all the formulas are already in the excel sheet and will calculate the relative expression. Good luck
1 Recommendation
Justus Liebig University Giessen
The answer of Korcan seems wrong. You can not compare the relative amounts of two different genes based on dCt values. Here is why:
Exponential amplification model:
F = p * N0 * E^C
p is a gene-specific proportionality factor. At the threshold F=Ft the cycle is C=Ct:
Ft = p * N0 * E^Ct
taking the log to base E:
log(Ft) = log(p) + log(N0) + Ct
Ct = -log(N0) + log(Ft) - log(p)
Consider this for two genes, A and B with initial concentartions A0 and B0:
Ct[A] = -log(A0) + log(Ft[A]) - log(p[A])
Ct[B] = -log(B0) + log(Ft[B]) - log(p[B])
the dCt is Ct[B] - Ct[A]:
dCt = -log(B0) + log(Ft[B]) - log(p[B]) + log(A0) - log(Ft[A]) + log(p[A])
dCt = log(A0)-log(B0) + log(Ft[B])-log(Ft[A]) + log(p[A])-log(p[B])
dCt = log(A0/B0) + log(Ft[B]/Ft[A]) + log(p[A]/p[B])
Now you may measure both genes at the same threshold so that Ft[A]=Ft[B] and thus
dCt = log(A0/B0) + log(p[A]/p[B])
Still the proportionality factors are part of the equaltion and the do not cancel out. This means that the dCt is proportional to the (log-)ratio, but with an unknown proportionality factor. I think this is not what was required.
The only way out is to determine these strange proportionality factors. This is done implicitely in an absolute quantification. Therefore I suggest to perform an absolute quantification of the two genes.
NB: the proportionality thing cancels out in the ddCt - but this was not the question here!
1 Recommendation
Gachon University
Dear Korcan, Anna and Jochen,
First of all thank you so much for your answer.
Korcan and Anna: How did you think about the Jochen comment ?
Jochen: If so, could you help me with the formula for calculate the relative ratio of the genes by qRT-PCR by another value; because I found many paper they also can use qRT-PCR to check the basal mRNA relative expression.
I hope to receive your answer as soon as possible.
Thank you so much for your support !
Warm regard !
Justus Liebig University Giessen
As I said: perform an absolute quantification of the two genes using an external calibrator.
Again, just to clarify:
You can compare the relative expression of one particular mRNA between different conditions or cell types. You can do this by simply comparing the dCt values (-> ddCt). I think this is what most papers do.
A comparison between different genes (in the same condition or cell type) is not that simple. Comparing dCt values for different target genes is meaningless (well, if the differences are huge (>>10x), then a comparison will likely point in the right direction; but then only the direction is indicated, so the conclusion is qualitative, not quantitative!).
Note that the dCt value is a normalized value of a target gene. The normalizer is a (mean) ct value from one or more reference gene(s) which expression(s) serve as a loading control. A single dCt value does not tell you much. It is just a value "in outer space", without a sensible reference point. In particular, a dCt of zero does *NOT* indicate that the expression of the target gene is as high as the expression of the reference gene. It is also irrelevant or uninformative if the dCt is positive or negative. The dCt contains this unknown proportionality factor. BUT: different dCt values for the same target gene (and the same reference gene(s)) can be compared, because they contain the same (unknown) proportionality factors that cancel out in the difference of the dCt values (i.e. in the ddCt value).
If your problem is setting the threshold value:
You can place the threshold at any value where the amplification curves show an exponential increase (and where all these log-linear segments are parallel).
Shifting the threshold will shift the Ct-values. But because the curves are parallel, all Ct values will be shifted by exactly the same amount. Hence, the dCt values will not be influenced by the threshold.
As you can see from the formulas I posted previousely, there is no need to chose the same threshold for the traget and the reference gene. The proportionality factors will cancel out in the ddCt as long as the same threshold (per gene) is used for the different conditions [or the same threshold (per condition) is used for the different genes {target and reference} - what is mathematically ok but rarely sensible in practice].
3 Recommendations
As far as I know from many post I have read here, Jochen is an expert on this and he is right. If you want to publish this comparison data or use it as an argument for your hypothesis, you have to use absolute quantification.
But if you only want to have an idea of relative expression of a group of related genes (for example, you are working on a cytokine and there are 2 receptors for this cytokine, you want to find out which is the prominent receptor for your cell type/state) in this case you can do relative quantification. I always do this before ordering antibodies, antagonists, inhibitors etc.
Similar questions and discussions
Related Publications
Neolamarckia cadamba is an economically-important fast-growing tree species in South China and Southeast Asia. As a prerequisite first step for future gene expression studies, we have identified and characterized a series of stable reference genes that can be used as controls for quantitative real time PCR (qRT-PCR) expression analysis in this stud...
The mRNA levels of S100A6.
The Relative mRNA Expression(2-ΔΔCt) levels of S100A6 were measured by qRT-PCR.
(XLSX)
Human papilloma virus (HPV) may cause mostly transient infections of cutaneous and mucous epithelia. Persistent HPV genital infections may induce pre-malignant or malignant lesions. While E6 and E7 HPV genes' malignant character is known, E5 is still under debate. We evaluated the possible role of E5 gene in cervix oncogenesis, in patients with abn...