I have done qPCR on several genes along with the reference gene for mutant and wild type samples. I have got the Ct values. Which is the best free software where I can feed the Ct values and get graphs (like we see in publications) with error bars using ∆∆Ct method? There is a way to do it in excel but it is little time consuming as I have several genes. I was looking for some online tool where we can just put the Ct values and graphs are generated with standard deviations with reference to the endogenous control across the two sets of samples. Please let me know. Thanks.
Dear Researcher
have you used qpcr software?then what about dataAssist software.you can use it.
Regards
Dear Sachin
i can send you an excel spreadsheet tomorrow with a worked example of delta delta Ct. You can then simply substitute in your own values. Shouldn't be too much work
Thanks Laurence, that will be of great help. I have about 35 genes and one reference. Then I kept 3 biological replicates and 3 technical, over all 9 reactions per gene per sample. Then some values are deviating a lot, so I have to omit them. All I want is a software or a spreadsheet like yours, where I can put the values of the reference and my target genes and make graphs with standard error. Thanks once again.
Dear Sachin
Find below an example of delta delta Ct analysis: You can simply plus your value into the relevant fields and it will change graphed values accordingly.
You will probably need to replicate cells etc based on the particualrs of your experimental design but replicated cells preserve embedded formulas so shouldn't be too arduous
Note: Regarding the above comment relative expression is calculated (as you can see) by using normalised data for each sample set, i.e. test and control minus housekeeper as an exponent for '2'. In other words this presupposes 'perfect doubling' for your housekeeper and GOI in each sample set and for that relationship to be valid a prior standard curve should demonstrate a primer efficiency of (ideally) 95%-110% (although I tend to find this method acceptable for efficiencies >/= 85%)
Below that value you modify the so called Delta delta ct value by introducing a co efficient pertaining to the actual primer efficiency: For example (2 x 'x') to the power of 'y' where 'x' = the actual primer replication efficiency, e.g. 80% =0.8 and y is the ther normalised integer used as an exponent: This is the basis of the Pfaffl method which is scripted by REST
Hi Laurence, I have more reaction than yours, 9 for each sample (3 biological X 3 technical). I will have to adjust the sheet according to it. Thanks.
Hi Sachin
yes if you copy and past cells that will preserve embedded formulas. I do that routinely when I have different plate designs
Dear Sachin
if values computed by scripted software such as REST are comparable to values in my spread sheet - which is the so called Delta Delta Ct or Livak method of analysis - +/- SEM - then they are effectively the same
Note that the Livak Method - and hence my spread sheet - pre supposes a primer amplification efficiency of > 85% (and ideally 95% -105%). You need to verify this by standard curve
The REST software I note is based on Pfaffl method of computing expression which actually factors in the exact efficiency of amplification (and does not pre suppose a perfect doubling per cycle, therefore utilising a co efficient of 2' in the expression calculation; as per my spread sheet)
I think the 2 methods are effectively the same, i.e. CT +/- SEM where primer efficiencies are > 85%; For lower primer efficiencies it is generally prudent re design your primers !!
Dear Lawrence,
I have used a similar sheet before, we used ∆∆Ct as zero for only one of the samples samples out of 6, like WT1 value for ∆∆Ct is zero, then WT-2, Wt-3, Mu-1, Mu-2 and Mu-3. Like in your case H8-H8 is zero, but for other samples we take that as zero and then do like that H9-H8, H10-H8, H11-H8 and so on for all samples. Please see attached, we take mean of WT1, 2 and 3 and then make a graph. Does this look OK?
Hello Sachin
I apologise for the much delayed response; I don't always check this forum and what with the Easter break I havn't looked at RG in more than 1 week
Anyways, I have looked at your spread sheet and I basically agree with your logic with one caveat: When dealing with separate samples, e.g. Mu-1, 2 and 3 you should compute values to plot of graphs based on the GEOMETRIC mean and not the ARITHMETIC mean: The latter can be used when you are performing calculations on Ct values derived from the same cDNA, i.e. technical replicates but NOT when performing calculations derived from separate cDNA samples, i.e. biological replicates
Im actually not sure whether Mu-1 2 and 3 data sets are derived from the same cDNA, i.e. where the arithmetic mean applies; in which case your graph is completely acceptable; or PCR performed on separate cDNA samples, i.e. biological replicates in which case plots of the geometric mean apply
Regardless, in your case if you look at your (attached & amended) spreadsheet where I have computed alternative geometric means and then used your graph to create a 2nd graph above yours they are effectively (i.e. within error limits) the same so your data has in effect been independently validated as good !!
The mathematical explanation for distinctions between the 2 calculations is provided in the link below:
https://www.algebra.com/algebra/homework/Average/Geometric-Mean.lesson
http://www.investopedia.com/ask/answers/06/geometricmean.asp
Hi Laurence, Mu1, 2 and 3 and also WT 1, 2 and 3 are separate biological replicates. For each biological replicate I took 3 technical replicates. I also fed these values in dataassist software which gives the fold change in RQ values. I am not sure how that is different. RQ value then RQ minimum and RQ maximum.
- Badges
- Science topic