Hi all,

I'm new to the field, so I need a little bit of help.

I have GC-MC results from samples of gland secretions and I want to test for individual repeatability and inter-individual similarity. I looked into the literature but in the majority of studies, they either use a subsample of peaks, use the principal components obtained from the PCA for the comparison, or use measures that just account for the variance in the sample. Is there a way to obtain an accurate and comprehensive (i.e. taking into account and comparing all the peaks) measure of similarity that can be used? Ideally, I would like to obtain a matrix (with a pairwise comparison among all the individuals/samples tested) out of it too.

N.B. Just a limited number of the compounds in the samples have been identified, so I'm using raw data about retention time and (relative) area of the peaks. Moreover (of course) I don't have the same peaks in all the various samples, so I don't think the "usual" repeatability tests for paired data are an option (unless I modify the dataset a lot, that can be done but I don't know whether or not it's the right way to proceed).