Question - ANCOM

User Support
1

Hello,

I have a question regarding the output of ANCOM. Is there a way of getting the results by #OTU ID instead of species name?

My problem is that I have different OTUs for the same bacterial specie. Example:

#OTU ID	Taxa
6db820962a5bf3ba7e0fe7f3f9dcc994	D_0__Bacteria;D_1__Fusobacteria;D_2__Fusobacteriia;D_3__Fusobacteriales;D_4__Fusobacteriaceae;D_5__Fusobacterium;D_6__Fusobacterium necrophorum subsp. necrophorum
3480ddadc4fab3ef678d1b60d668f31a	D_0__Bacteria;D_1__Fusobacteria;D_2__Fusobacteriia;D_3__Fusobacteriales;D_4__Fusobacteriaceae;D_5__Fusobacterium;D_6__Fusobacterium necrophorum subsp. necrophorum
88ad345b22f4a093de81d3797cc83b19	D_0__Bacteria;D_1__Fusobacteria;D_2__Fusobacteriia;D_3__Fusobacteriales;D_4__Fusobacteriaceae;D_5__Fusobacterium;D_6__Fusobacterium necrophorum subsp. necrophorum
12f97dd6585cb81c36341be917a0ae85	D_0__Bacteria;D_1__Fusobacteria;D_2__Fusobacteriia;D_3__Fusobacteriales;D_4__Fusobacteriaceae;D_5__Fusobacterium;D_6__Fusobacterium necrophorum subsp. necrophorum
9373e92aaf629ef9bbce3f77878d6247	D_0__Bacteria;D_1__Fusobacteria;D_2__Fusobacteriia;D_3__Fusobacteriales;D_4__Fusobacteriaceae;D_5__Fusobacterium;D_6__Fusobacterium necrophorum subsp. necrophorum
c341b166936e40992db08d4da4ee890a	D_0__Bacteria;D_1__Fusobacteria;D_2__Fusobacteriia;D_3__Fusobacteriales;D_4__Fusobacteriaceae;D_5__Fusobacterium;D_6__Fusobacterium necrophorum subsp. necrophorum
d2b16eba14c6bfcd8d3a5d6db6a544a7	D_0__Bacteria;D_1__Fusobacteria;D_2__Fusobacteriia;D_3__Fusobacteriales;D_4__Fusobacteriaceae;D_5__Fusobacterium;D_6__Fusobacterium necrophorum subsp. necrophorum
c5b0ef668bb524e7e5f05c5bae818dec	D_0__Bacteria;D_1__Fusobacteria;D_2__Fusobacteriia;D_3__Fusobacteriales;D_4__Fusobacteriaceae;D_5__Fusobacterium;D_6__Fusobacterium necrophorum subsp. necrophorum
e5ab39a55950bad5d793606ae4f01942	D_0__Bacteria;D_1__Fusobacteria;D_2__Fusobacteriia;D_3__Fusobacteriales;D_4__Fusobacteriaceae;D_5__Fusobacterium;D_6__Fusobacterium necrophorum subsp. necrophorum

Thus, if I could get the ANCOM results by OTU ID instead of species name, that would be very helpful.
Please see the attached file containing my current results. result-ANCOM-DAY7.qzv (405.0 KB)

Thank you, Joao

2

Yes.
Check the tutorial
https://docs.qiime2.org/2020.2/tutorials/moving-pictures/#differential-abundance-testing-with-ancom
If you collapse your table, you will receive the output with taxonomy, but with ASV/OTU table (that is not collapsed) you will get ASVs or OTUs

2 Likes
3

Hi @JoaoGabrielMoraes,

Building on @timanix’s excellent answer, it looks like you may have imported an already collapsed table. If you run qiime feature-table summarize, what are your feature IDs?

Best,
Justine

1 Like
4

Thank you @timanix and @jwdebelius!!

I did not run qiime taxa collapse prior to ANCOM.

When I run qiime feature-table summarize on the input feature table used for the ANCOM analysis, and check the Feature Detail tab of the output file, I see that my features are shown as taxonomy. However, they are unique (I don’t see repeated taxonomy classification).

Does that mean that in my results of ANCOM, the OTUs with the exact same taxonomy classification were combined together?

Thank you, Joao

5

Hi @JoaoGabrielMoraes,

Your issue is happening with the import, so it depends on how your data was prepared before it came into QIIME. (I can only tell you want happened after!) did you maybe import a collapsed table? Are your taxonomic identifiers the first column? I would double check both those things.

Best,
Justine

1 Like
6

Thanks @jwdebelius! I am sorry, I am very new to microbiome research and QIIME, so I am only now beginning to understand it.

My data was initially processed by the DNA Core that generated our 16S sequencing data, so I am not sure about what was done prior to QIIME. I just got the feature table from them to run ANCOM.

What do you mean by taxonomic identifiers? Is it the names that I am getting with the taxonomy classification? If so, that’s the last column of my OTU table.

Thank you, Joao

7

Hi @JoaoGabrielMoraes,

Okay, my general advice is to always know how your data was generated! (And get the data as raw as possible). I assume you got a text file that you then converted to a biom table. Did you do that conversion, or did the sequencing core? You may need to talk to them about what happened to the data before you got it.

I would also just double check your imported files, particularly if you got multiple tables. I can’t count the number of times I grabbed the wrong file.

Best,
Justine

2 Likes
8

Thank you @jwdebelius!!

Following your directions I got the FASTQ files and will process the data from the beginning.

Thanks again, Joao

9

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