When >2 samples are analysed using the DMRforPairs parent function, the Kruskal-Wallis test is first applied and, if p<0.05, the Mann-Whitney test is applied pairwise. Whilst this provides a guideline for which samples differ, it isn't strictly valid according to these forum discussions: Are pairwise Wilcoxon tests a valid non-parametric alternative to Tukey's HSD test?, Kruskal-Wallis Test and Mann-Whitney U Test. The discussion recommends applying the Dunn test instead (though I have never used this myself).

Otherwise all working well! Thanks for a nice package.