Hello,
In the Quantitative Enrichment Analysis result table (img1 below): each time I ran the analysis the raw p-values seem to be inversely related to the q-values. How should I interpret this? And which results should I consider significant? E.g. if I go with the standard for raw p-value (<0.05), I'm getting very high (significantly above expected) q-value; if I go with a low q-value, p-value is beyond anything considered to be significant? Could this indicate problem with the data?
Also, for the box plots showing relative abundance of the hit metabolites: all of mine have values above and 0 zero. Should I interpret that as above normal (i.e. reference) abundance and below normal abundance (img2 below)?
Thank you!
Here's an example:
Note: it seems I have confused Q statistic with the Q- value; please disregard the 1st part of the question