Dear colleagues,
I recently jumped in this field and I was trying to recreate a study, using absolute qPCR quantification for comparison between the abundance of certain bacterial specie in two groups of samples. After optimization of my protocol, I ran 2-step qPCR on 96-well plates with samples in triplicates and I ran the reaction three times in order to get the variance.
I am going for absolute quantification, with a standard curve with Ef ranging from 1,979 to 1,891 (98,95 to 94,5 %) with LunaMix from NEB. However, even though I used the same protocol, my values have a big variance between the plates. I would like to normalize them to use them in the same data set in order to have a more robust analysis. The ratio between the diseased and healthy group (16/8 at each 95well plate) is the same at each plate.
Is there a way to normalize the data in order to use them in one dataset? Also, if you are just looking, it is necessary to have more than 1 reaction apart from technical replicates?
I will much appreciate your help.
Sincerely, Martin
Dear Martin Černý for me it is a good idea to do triplicates for each sample in one 96-well plate. In general it is enough to analize your data with a good precision. But, be carefull with adding of samples in reaction mix. It is vety difficult to control tip deppness in the sample tube and yiu can add more or less due to tip position differences.
For precision please use automated robotic station.
When you will use data from 3 different runs and that data is differ, you can lose precision. You will have a big "non signifacant" area.
Dear Dr. Babenka, thank you for your answer. Due to my budget, I have no possibility to use robotic station, so i have to do it manually, but as you know it does come with its flaws. So, do you think, the one plate with all samples and standards should be sufficient for analysis?
Hi,
yes as you run standards on the plates you should be good to go. One ting you should have in mind is that the sample quality/ potnetial PCR inhibitors might affect your quantification as I expect your standards are purified DNA. If you faces this issue you could consider to use dedicated mixes or internal controls. Let me know if this is interesting then we can elaborate more on this
Sven
- Badges
- Science method