Mothur can't find temporary files with classify.seqs

[rpeguilhan]

Hello,
I am currently trying to run the Miseq SOP exemple with some of my samples for a practical course I’m preparing.

I’m working on my lab’s computing cluster and the files required for MiSeq SOP (silva.bacteria and trainset9 files) are downloaded to a specific directory which can’t be my working directory.

Anyway, everything works up to the “classify.seqs” command line:

mothur > classify.seqs(fasta=/home/user/r/rpeguilhan/test/stability.trim.contigs.good.unique.good.filter.unique.precluster.denovo.vsearch.fasta, count=/home/user/r/rpeguilhan/test/stability.trim.contigs.good.unique.good.filter.unique.precluster.denovo.vsearch.count_table, reference=/opt/cluster/contrib/apps/anaconda/3/envs/trainset9_032012.pds.fasta, taxonomy=/opt/cluster/contrib/apps/anaconda/3/envs/trainset9_032012.pds.tax)

Using 12 processors.
Generating search database...    [ERROR]: Could not open /opt/cluster/contrib/apps/anaconda/3/envs/trainset9_032012.pds.8mer
DONE.
It took 11 seconds generate search database.

Reading in the /opt/cluster/contrib/apps/anaconda/3/envs/trainset9_032012.pds.tax taxonomy...	[ERROR]: Could not open /opt/cluster/contrib/apps/anaconda/3/envs/trainset9_032012.pds.tree.sum
DONE.
Calculating template taxonomy tree...     [ERROR]: Could not open /opt/cluster/contrib/apps/anaconda/3/envs/trainset9_032012.pds.tree.train
DONE.
Calculating template probabilities...     [ERROR]: Could not open /opt/cluster/contrib/apps/anaconda/3/envs/trainset9_032012.pds.trainset9_032012.pds.8mer.prob
[ERROR]: Could not open /opt/cluster/contrib/apps/anaconda/3/envs/trainset9_032012.pds.trainset9_032012.pds.8mer.numNonZero
DONE.
It took 17 seconds get probabilities.
Classifying sequences from /home/user/r/rpeguilhan/test/stability.trim.contigs.good.unique.good.filter.unique.precluster.denovo.vsearch.fasta ...
Mismatch with taxonomy and template files. Cannot continue.
Mismatch with taxonomy and template files. Cannot continue.

I tried to change my input and output directories with the command line set.dir but nothings helped. It seems that mothur can’t find the temporary files… but I don’t know what to do.

Can anyone help me?

Thank you.

[pschloss]

Rather than putting your reference files in your anaconda env directory, can you put them in /home/user/r/rpeguilhan/test/ instead? I suspect that’s were they are already. If that’s the case then you need to use the same path as you are using for your sequence data.

Pat