search

BeeCSI-Microbiome / R_analyses

1 stars 0 forks source link

ANCOMBC doesn't correctly adjust p-value when there is more than 1 contrast (i.e. 3 or more treatments) #20

Open LLansing opened 2 years ago

LLansing commented 2 years ago

ANCOMBC doesn't correctly adjust p-value when there is more than 1 contrast (i.e. 3 or more treatments), according to this comment on an ANCOMBC issue by one of its devs.

They also mention the following:

We are incorporating mixed differential false discovery rate (mdFDR) control into ancombc to solve this problem. The new update is likely to be pushed by next month.

So we should review the procedure when the update is published.

JeremyTournayre commented 1 year ago

Hello,

If I understand correctly your message which the issue "https://github.com/FrederickHuangLin/ANCOMBC/issues/64" is cited, you can do all comparisons 2 by 2 with ANCOM BC ? Because I use relevel to change the reference group to do all comparisons 2 by 2 but maybe there is another solution since updates?

Best, Jérémy Tournayre