I’m trying to find the significance level used when running ANCOM. While the W statistics helps to identify taxa that are differentially abundant, I don’t see a way to report either a p-value or at least a significance threshold. I’ve found on the developer’s website examples that use a significance level of 0.05, but don’t see in QIIME which significance threshold is used.
Right, it isn’t possible to report a p-value, because there aren’t any.
This is a feature, not a bug, since it isn’t possible to generate reliable p-values in the context of differential abundance in the first place.
The significance level for the qiime implementation is also 0.05 - we just didn’t expose it to the command line interface since we are just using all of the default parameters. Pull requests are welcome to expose these parameters – it’ll just require modifying this line of code and adding extra command line arguments here
2 Likes
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.