The “16S Microbial analysis with Mothur” tutorial requires a taxonomy file in Classify.seqs. The taxonomy file is supplied (trainset9032012.pds.tax), but it gives an error “No datasets with mothur.seq.taxonomy elements available”. I tried newer tax files and checked the help, but no answer. Any ideas? Thanks.
Welcome @bburnes
The error message is probably referring to the other inputs.
Tutorial step Hands-on: 16S Microbial Analysis with mothur (extended) / 16S Microbial Analysis with mothur (extended) / Microbiome
Screenshot
Did you source both the fasta and tax datasets from the tutorial? Then, are both the fasta and count table inputs from the prior Remove.seqs step?
I can let you know that the taxonomy dataset has a special format. It is important to load this up to Galaxy using all defaults in the Upload tool. Then, allow Galaxy to perform the implicit format conversion during runtime – there will be another copy of the file in your history nested inside of the original file (same data, different format). Try to not disturb that data. If you already did, try loading the data again this way to see if it can resolve the current problem.
Then, tools in this protocol are used a few times, so it is pretty easy to mix up the datasets! Later on you can use a workflow that will do the data sorting for you.
Finally, I just started up the tutorial’s workflow at UseGalaxy.org to see what happens! This is the shared workflow invocation.
Please let us know if you are able to resolve this or not! You are also welcome to share back your history for a closer review. 
How to get faster help with your question
Hi Jenna,
Thanks for replying so fast. To answer your questions:
Did you source both the fasta and tax datasets from the tutorial? Yes, but then I realized that there were much newer versions of those datasets, so I tried to use them. I copied the workflow you shared, clicked run, then tried to enter the newer datasets. But I got the same message where it wants trainset9_032012.pds.tax…it says “No datasets with mothur.seq.taxonomy elements available”. Looking online, it says the mothur.seq.taxonomy file is generated from the classify.seqs step, but that is the step I am trying to run that requires the mothur.seq.taxonomy file.
Then, are both the fasta and count table inputs from the prior Remove.seqs step? Yes.
Thanks, again.
Brian
Thanks for explaining more @bburnes !
I would suggest trying with the tutorial’s version of these datasets first, if only so you can get through the remainder of the protocol. Or, is this what you have already done?
The job in my test history above appears to have worked.
There are a few items here to notice, including the format of the reference data.
Then, for the updated database versions you sourced, I’m guessing that there is either a format issue or some mismatch between the tax and fasta. The primary “key” in both files is the same – this is how the tool links them together. For the fasta, the key is the >identifier and in the tax, the key is in the first column of the tabular file.
And, the input datasets from Remove.seqs has a similar organization: the same number of entries plus a common key value between the fasta and tabular file.
Maybe this helps to review the job content? You are still welcome to share back your history for feedback. If there is some problem with the tutorial protocol we would like to help to solve it. Maybe we can add in another tip to the tutorial to help others avoid the same issue!
I ran it from the tutorial link and it worked fine. Then I subbed in my data and got the same error. Then I tried again, but waited a while after uploading my datasets and it worked! You were right, it took a minute to process the .tax file behind the scene before it would accept it. Thanks so much!!!
And I love your buterfly icon.
Brian
1 Like
Great, glad everything is working now! Thanks for letting us know. 