Remove dependency on "microbiome"

[mlbendall]

The core microbiome functions depend on functions in https://github.com/microbiome/microbiome, which introduces a lot of additional dependencies.

[ecastron]

Perhaps it's not so convenient to add that one then. It's a nice to have feature, and you don't find many packages to do that type of analysis, but if it's going to burden too much pathoStat we could leave it for another future version.

[mlbendall]

I don't think it'll be too hard to remove, they're basically just adjusting detection/prevalence cutoffs, right? I'm just reminding myself to do it.

[ecastron]

I think so, it's basically core taxa detection as a function of abundance with different prevalences

[mlbendall]

I think the only tricky part is figuring out what those thresholds should be, because if you set it up wrong you get something like seen below. Although I'm still trying to figure out if this is because the example data is pre-filtered (for top 10 taxa) or because of the default thresholds.

[ecastron]

It looks like they use 1 and 95% for detection and prevalence threshold, respectively. In pan genome analysis they use the same prevalence threshold for genes, i.e., being present in 95% of the genomes