Hello out there
I've been looking at the expression of 15 different genes in two strains of lactic acid bacteria after they have been exposed to Staphylococcus in a simulated co-culture after 5, 10, and 20 hours of incubation. As a control, the two strains of lactic acid bacteria were grown in a normal medium. Both groups were carried out in triplicates.
I've calculated the gene expression relative to GyrA using the Pfaffl method, which works quite well.
However, I'm confused about which kind of statistic test is most appropriate to analyze my qPCR data? From what I can understand, I need to test my sample (the co-culture) against the control (normal culture) for each gene at each time point, meaning that I have a sample size of 3 per gene. Firstly, I did a Shapiro-Wilk test to ensure normality before performing a t-test (Welch's). I've now been made aware that I can't use any of these tests when my sample size is this small and that I can't assume any normality.
I'm now puzzled about how to address the statistic analysis. I've tried to look for a non-parametric test and I came by the Mann-Whitney test, which I believe is the pendant to the unpaired t-test. Yet, none of my results har significant when using this test even though I can see a 4-fold change in the expression of certain genes.
Does anyone have some ideas?
Thanks in advance!
- Badges
- Science method