I am having trouble wrapping my head around and describing the meaning of the changes in my emperor plot after adding the custom axis option. I am comparing microbial communities from samples collected from the same individuals pre and post treatment. I have colored the points on my plot based on the categorical variable of "Pre" or "Post", in order to make a custom axis using a continuous variable, I changed the pre and post to numbers representing the day since the start of the experiment. I have attached screenshots of the original emperor plot, the same plot utilizing a custom axis, and the results of a PERMANOVA test for significance on the pre vs. post metadata column.
I would greatly appreciate if someone could explain to me how the changes in the clustering of the plots are generated and what this means in a broader view of the biological trends present.
I would greatly appreciate if someone could explain to me how the changes in the clustering of the plots are generated and what this means in a broader view of the biological trends present.
The changes you see are a consequence of the DaysSinceExperimentStart values. When this option is used all samples with the same value under this category will be colocated along what usually would be Axis 1. Note that these changes are only intended as a visual aid for the ordination.
PERMANOVA relies on the distance matrix i.e. all the pairwise distances between the samples. As such the custom axis post-processing has no effect in this test.
