I have a triplicate of 2 groups from a SILAC experiment. As I want to produce a volcano plot, I'm trying to obtain the p-value for each protein using a T-test, using scipy (scipy.stats.ttest_ind_from_stats). The module asks for 3 parameters for each group, the median, the standard deviation, and the number of observations. My question is, what is the best parameter for the number of observations? The number of replicates of the experiment? Or the peptide counts for each protein?
Also, is it better using the log2(ratio) median and standard deviation for the other two parameters, or the normalized abundance (median and std) is fine?
Thank you all for your time!