I have 3 experimental conditions, as well as a control. Each of these 4 conditions has 3 biological replicates. When performing the qPCR, each biological replicate was run in triplicate (technical replicates). My qPCR experiment was simple: looking at a gene of interest, compared to a single housekeeping gene. Using the delta-delta CT method, at what stage should I be averaging for each condition? How should I be calculating the standard deviation?
Would I average all the technical replicate CT values for COND1.1, and treat that as a distinct condition? Or, would I average all the technical and biological replicate CT values for COND1? Would I average the technical replicate CT values for COND1.1 at the beginning, and then average our their final expression values? But then, how would I decide which biological replicate of the CONTROL would be the normaliser, without introducing bias? At which stage with the final presented SD be calculated?
I have analysed the data about 4 times now, with my lab unable to come to a consensus on these questions, so I am hoping to gather further insight.
Thank you!
Hello Rhiannon,
This is one of those topics that I think it is explained better graphically, than with words, so I have made two figures to accompany the explanation (see attached files).
First of all, you need to calculate the average delta Ct of your control (as explained in "I. Control.tif"), and then, use it to calculate delta-delta Ct of your conditions of interest (as explained in "II. Conditions 1, 2 and 3.tif"). I recommend you to use statistical analysis programs such as GraphPad Prism, which allows you to enter the individual delta-delta Ct values, in addition to calculating the mean with SD or SEM, and compare it with other conditions in the same analysis.
Also, for the analysis to be consistent and meaningful, keep in mind that the individual Ct values of the technical replicates should not differ by more than 1 - 1.5 from each other.
I hope I have been helpful, have a good day.
It is possible you find something information in the web site Applied Biosystem.
Hello Rhiannon Jamieson-Williams
In case of me I have conducted q-pcr using house keeping gene B-actin. For three replication of biological experiments. I have done 3 times q-pcr for three replication. Then I think you will got delta-delta CT result according to b-actin. After that getting the relative quantity value. You need to make average of that value and then you can also calculate standard deviation of that 3 value.
Other wise after getting the delta-delta CT result. If you have negative control like in case of cell line culture (negative control). Now need to calculate again according to negative control. Need to divide the each value of each experiment by each negative control value. Then you can get value for each experiment then make average of 3 experiments and calculate standard deviation of this 3 value.
Sorry for my answer, but you should not use the delta delta method anymore. This method presumes equal efficiencies for all targets under study, whereas its' well known that efficiencies can vary in the most extreme case from 1.7-2.0 even for optimized targets, due to target-primer specific characteristics even after optimization. Moreover many analytical parametrs influence the efficiency as pipetting and inhibitor errors do. With that in mind The Cq can vary more than 6 units within an experimental setting due to different slopes in individual amplification curves. At least use the Pfaffl method with a correction factor for different efficiences in the amplification of your targets. Better is to calculate the N0 for each individual amplification curve and analyse your gene expression with LinReg PCR. If you are interested I will give the references.
Hello Rhiannon,
This is one of those topics that I think it is explained better graphically, than with words, so I have made two figures to accompany the explanation (see attached files).
First of all, you need to calculate the average delta Ct of your control (as explained in "I. Control.tif"), and then, use it to calculate delta-delta Ct of your conditions of interest (as explained in "II. Conditions 1, 2 and 3.tif"). I recommend you to use statistical analysis programs such as GraphPad Prism, which allows you to enter the individual delta-delta Ct values, in addition to calculating the mean with SD or SEM, and compare it with other conditions in the same analysis.
Also, for the analysis to be consistent and meaningful, keep in mind that the individual Ct values of the technical replicates should not differ by more than 1 - 1.5 from each other.
I hope I have been helpful, have a good day.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics