ANCOM F-statistic detail

Dear All,

I am now applying ANCOM to find the distinct taxa between my samples. When I get the result, I am confusing about the detail of "clr" in the visualisation plot, like below:

I have search the forum and I guess the meaning of the x-axis is something like "fold change", to show how different of the taxa changed between groups. But I want to know how exactly to calculate the "clr".

More specifically, in the plot I shown, the red point has a W score of 20 and clr of 2.8. How does this "2.8" come from?

Thanks for all the ideas

1 Like

Hello @Lennon_Lee,

That’s a great questions. As far as I can tell, the clr value shown here is the results of the centre log ratio transformation that’s part of skbio.stats.composition.clr.

Of course, that doesn’t answer the question, “Why would I want to know the center log ratio of this data”.

Let’s see if @mortonjt can offer some advice on making use of this CLR value.


Happy New Year! :clinking_glasses:

Thanks for the reply. Actually I want to plot the graph by my own in R, but I found that I couldn’t export the data from “Export as TSV”. Anyway, thanks for your help and I can calculate by mu own, happy new year !!

Hi @Lennon_Lee, this is a known bug and should be fixed soon (possibly the next release of QIIME 2, occurring in January, but I don’t want to make promises :wink:).

1 Like

Noted with thanks!:hugs:

1 Like

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.