Problems with the taxonomy file: Different tree taxonomy of Bacteroides at genus level, and duplicate results for different OTU.IDs

[Ehud]

Hi,
I am not sure that User Support is the right place for this question, but I don’t which forum is better for this.
The problem is I get a taxonomic file with some strange behavior and wonder what the meaning of this issues.
first: I get Bacteroides at genus level with 2 different tree taxonomy. one is the right Family according to literature and I think that the second result is mistake. attached 2 examples from the taxonomy file genarated by qiime2:
OTU.ID Taxon Confidence
9a0313e1e0e7cfcc434816ea9e1657a9 k__Bacteria; p__Bacteroidetes; c__Bacteroidia; o__Bacteroidales; f__Bacteroidaceae; g__Bacteroides; s__uniformis 0.7789094
6bb28ac1ad77fd6c4defab826c8fd81e k__Bacteria; p__Firmicutes; c__Clostridia; o__Clostridiales; f__Lachnospiraceae; g__Bacteroides; s__xylanolyticus 0.7313287

How to fix that? and how i can be sure that no more of such this mistakes?

second: I have a lot of different OTU.IDs with exactly the same full taxonomy. attached an example:
OTU.ID Taxon Confidence
20a60029c686b5b3100970930ec19760 k__Bacteria; p__Bacteroidetes; c__Bacteroidia; o__Bacteroidales; f__Bacteroidaceae; g__Bacteroides; s__uniformis 0.9072604
952ad49052673a9585fb960d3ad0ce22 k__Bacteria; p__Bacteroidetes; c__Bacteroidia; o__Bacteroidales; f__Bacteroidaceae; g__Bacteroides; s__uniformis 0.9727496
e9b5d0ae3b3e56a939220cc5502456b8 k__Bacteria; p__Bacteroidetes; c__Bacteroidia; o__Bacteroidales; f__Bacteroidaceae; g__Bacteroides; s__uniformis 0.9971192

Are this possible? How this samples count when I collapse the data to table with specific level?

Thank you very much
Ehud

[Nicholas_Bokulich]

@Ehud

first: I get Bacteroides at genus level with 2 different tree taxonomy. one is the right Family according to literature and I think that the second result is mistake. attached 2 examples from the taxonomy file genarated by qiime2:

This is a problem with your database, not with QIIME 2. You should contact whoever made the database to ask them why you have a Bacteroides in the wrong phylum.

@Ehud

second: I have a lot of different OTU.IDs with exactly the same full taxonomy. attached an example:

This is normal. Just because a feature has a unique amplicon sequence does not mean that it is a unique species. So it is very typical to see many features with the same taxonomy assignment.

@Ehud

How this samples count when I collapse the data to table with specific level?

Those features would collapse into one.

[Ehud]

Hi @Nicholas_Bokulich,
Thanks for your fast respond.

Nicholas_Bokulich:

This is a problem with your database, not with QIIME 2. You should contact whoever made the database to ask them why you have a Bacteroides in the wrong phylum.

Because I am really beginner with code I followed exactly after the ‘moving pictures’ tutorial. According to that I used the Greengenes 13_8 data base as mentioned there…

@Nicholas_Bokulich

This is normal. Just because a feature has a unique amplicon sequence does not mean that it is a unique species. So it is very typical to see many features with the same taxonomy assignment.

@Nicholas_Bokulich

Those features would collapse into one.

great!

Thanks a lot!
Ehud

[Nicholas_Bokulich]

@Ehud

Because I am really beginner with code I followed exactly after the ‘moving pictures’ tutorial. According to that I used the Greengenes 13_8 data base as mentioned there…

So contact the greengenes developers to ask about that sequence annotation. I expect there is a good explanation.

The QIIME 2 developers are not involved in the creation of this database. We only provide pre-trained classifiers made from various databases, but do not perform any sort of data curation on these. Ultimately, it is up the user (and database maintainers) to vouch for the quality of their (reference) data.