search

ggloor / ALDEx_bioc

ALDEx_bioc is the working directory for updating bioconductor
26 stars 13 forks source link

Feature request #9

Closed stro0070 closed 2 years ago

stro0070 commented 6 years ago

Hi,

Can ALDEx2 be expanded to do post-hoc testing and report results of comparisons between multiple groups, like it does between two groups? For example, my project has 16 different treatments, and I would like to be able to see whether features (OTUs) are significantly more abundant in certain treatments over others, identified by a Tukey's HSD significance category letter or something similar. It would also be great to be able to extract mean CLRs from each treatment as well as CLRs associated with features (OTUs) from individual samples.

Thanks, Noah Strom

145218415 commented 3 years ago

Hi,

I have the same question, there is any advance conserning this functionnality ?

In my case I got simple experimental design : 4 different treatments and only 1 timepoint at the end of the experiment, so I thought I should use aldex.kw(), and then apply a post-hoc test allowing comparison between treatments.

If I understood well, in order to be able to compare abundances between 2 groups, I have to use instead aldex.glm() with a formula like that : mm <- model.matrix(~ treatment + Mouse.ID , covariates) ?

Thank you in advance ! Thibault

ggloor commented 3 years ago

I have not yet included post-hoc tests, but I agree it is urgent to do so. And yes, the aldex.glm function is what you should use.

trichterhb commented 2 years ago

Hi, maybe a bit late to the party, but what is the status of the post-hoc? It would be extremely valuable to have.

ggloor commented 2 years ago

Long story short. I will be introducing a post-hoc test into the next version, but it will not be Tukey's HSD.

Tukey's test requires an ANOVA in the back end, and so the HSD test would be on something other than the glm result that is reported. Since Tukey's HSD is simply a kind of FWER correction, I will include the option of choosing the FWER correction. At the moment ALDEx2 uses the holm-bonferroni test (https://en.wikipedia.org/wiki/Holm–Bonferroni_method) which is most likely the best choice. However, the update will allow for FDR testing as well.

As for extracting mean clr values from each comparison, see aldex.glm.effect() with the include.sample.summary=T option

Hopefully this will be helpful