Leo Lahti

Thanks for the report. Without seeing the files it is difficult to debug exactly. My impression is that the m.otu file is not properly read or matched, and as a...

This may be related to the use of vapply vs. sapply. We are advised to switch to vapply but this causes problems currently. I tried to fix and change to...

Which version you are using? I am able to run your above example (with zip) using the latest devel version 1.9.14 of microbiome pkg.

Also updated from sapply to vapply in the latest devel version 1.9.15 now (in github master)

This should be ok in an online tutorial but for a package manpage example it should be in the package itself. This is not possible due to file size, hence...

Can you update to the latest version with devtools::install_github("microbiome/microbiome") and then report both microbiome and R version if the problem persists?

This is a vast question. In general, generalized linear models (negative binomial etc) are more appropriate for count data and then library size can be included as a variable, or...

The "group1" refers to the group name. In your data is something else "G1" so use that.

I can see the motivation. I realized, however, that the function is overly complex to maintain and gets easily unstable with any changes. I would propose splitting it into multiple...

Ok fine for now. But we may have to reconsider these policies when the overall project goes fwd and coherence is needed. If your function extends the old one and...