Hi Everyone,
I have a dataset from proteomics analysis of treated cancer cell line, I have three replicates in treatment time-points and two replicate in control condition.
Now, How can I calculate p-value for my dataset?
I tried perseus, MSstatTMT package in R and functions in excel, all of them report different p-values. which method is better or more recomended?
A t-test paired with unequal sampling sizes makes no sense, either conceptually or mathematically. Conceptually, if your "before" values differ a lot, then a paired t-test is good relative to the difference between your before and after values.
Maybe try non-parametric tests (e.g. Kruskal-Wallis test or Manny-White test). I've been in a similar situation in a past study (skewed distributions or group sizes) and found the Kruskal-Wallis test can be as good as an ANOVA.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Sampling Design
- Sample Size