ANCOM results: ASV with W values of 0 identified as differentially abundant

[JiaqianLi]

Hi everyone,
I tried to use ANCOM to identify the biomarker at ASV level.
There were lots of ASV did be indentified as diffrentially abunant, like piture below:

But the w value are very low even =0
Can I use those ASV as biomarker? Below is my volcano results.

Thanks a lot!

[colinbrislawn (Colin J Brislawn)]

Good morning,

But the w value are very low even =0

You may have found this already, but if not, check out this thread on ANCOM (1) and low W values:

ANCOM: 'low W taxa identified as significant' issue's workaround, ANCOM2 code/instructions

Hi folks, Some users (including myself) have noticed that in some cases ANCOM behaves oddly in that it identifies very low/0 W valued taxa as differentially abundant. See here and here for example of this behavior. In contacting one of the developer's of ANCOM regarding this issue ,Dr. Peddada had the following explanation/recommendation. A potential reason for the problem is that the earlier code was trying to empirically derive the threshold for significance. When W takes on small values …

Also check out the ANCOM-BC plugin for Qiime2:

GitHub - mortonjt/q2-ancombc: qiime2 plugin for ANCOMBC

[JiaqianLi]

Thanks for your detailed reply.
I tried both rarefied and unrarefied table. The results are similar but not identically same.

Do you have any recommendation on which one is more reliable? I am a little bit confusing because I read a previous reply stated that the results derived from unrarefied and rarefied table supposed to be same. Thanks a lot!

[colinbrislawn (Colin J Brislawn)]

ANCOM is designed to be used on unrarefied data. (It does the normalization itself!)

We would not expect rarefied and raw data to be identical, as rarefaction throws away data to get the same counts per sample. However, big trends should be the same regardless of normalization.

[JiaqianLi]

Got it! Thank you so much!