Hey there,
If you’re interested in differential abundance testing, I would suggest you try multiple approaches and make sure any inferences you make are robust to the choice of tool (i.e., that the functions that differ are identified by multiple DA tools).
You could also look at PCoAs, such as in the qiime2 tutorial on the plugin.
Last, STAMP is a useful tool for visualizing these datatypes, although I would not recommend relying on the statistical tests as they are not appropriate for compositional data.
Hopefully that helps,