Issue because of `collapse` version

Downstream analysis and statistics MaAsLin
1

Hello.

Many thanks for all the contribution!

This is just to let the community know of an observation I made.

Using MaAsLin 3 (maaslin3_1.5.5) in R.4.4.1, I noticed a funny behavior caused by the collapse library.

My R used to have collapse2.0.16, this would to successful runs that would have NA values across al the all_results.tsvfile and apparently, empty significant_results.tsv file.

After upgrading the collapse package to a version higher than 2.1.0 so the:

> "fslice" %in% getNamespaceExports("collapse")
TRUE

MaAsLin3 worked fine.

If you think I’m loosing something here, please let me know! :slight_smile:
Many thanks again!

2

Thanks for pointing this out. We’ve updated the dependencies to require collapse to be 2.1.0 or higher.

Will

1 Like