Currently, mostly our datasets are from human microbiome. Consider adding new datasets that have environmental samples (preferably multiomics). As these methods can be applied across different systems, more diverse set of example datasets would highlight this more (although switching datasets can also make the content more confusing). Also something to consider for September workshop. @antagomir