I have been doing a lot of qPCRs and one of my genes gets a lot of "No Ct value"s so a colleague suggested going back into the Aria software and changing it, so the two technical replicates are not automatically averaged but instead shown as two individual values to see if there was any values there. And for some there were which is great.
but my question is, is this a widely accepted things to do?
If you see frequent "No Ct" values for a gene, you should first test your reagents (does it work well in a positive control?). If your positive control works, you're probably at/below the detection limit for the assay. The Ct value you see may be "noise", especially if the Ct values are getting are inconsistent. qPCR can detect even single molecules of the target cDNA, but that one molecule can't be in two wells simultaneously.
You should definitely not use the other Ct value if one is undetermined, especially if you are only running the assay in duplicate. The presence of a Ct value is in no way "more correct" than an undetermined Ct value unless you can prove that there was a technical issue. Running the assay in triplicate can give you a bit more confidence in Ct values.
You should also not set an undetermined Ct to the maximum cycle number. If you need to compare between samples, you should report it as not detected in samples without a Ct. In this case, you can show the expression level in the samples where it was detected based on the ∆Ct (i.e., relative to your housekeeping gene(s)).
It is not uncommon to analyze technical replicates separately in qPCR experiments, particularly when some replicates give no Ct value. This can help to identify variability in the reaction and potentially salvage data that would have been lost otherwise. However, it is important to note that this approach may not be suitable for all qPCR experiments, and the decision to analyze technical replicates separately should be made based on the specific experimental design and research question.
If you do decide to analyze technical replicates separately, it is important to report this in your methods and results sections, as well as the reason for doing so. You should also consider any potential implications of this approach on statistical analysis and interpretation of your results, and consult with a statistician or bioinformatician if necessary.
Thank you so much for your advice John Hardy Lockhart running in triple replicate is a great idea for this gene that has such low expression in some samples
Ahmad Al Khraisat that is a really great point to include in my methods so that it's completely transparent. As this is more of a cool side project and not the sole focus of my master's project I think it's acceptable. In a perfect world with lots of time and money, I would spend time optimizing the assay.
Really appreciate your help
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Modeling
- Analyses of Variance
- ANOVA