summary.tax issue

mothur bugs
1

Hi. I am sorry: I opened an issue in your github repository, but I have seen that you probably prefer the forum to tackle with this staff. So, I paste here the link:

Many thanks,
Jose

2

Mothur is expecting WYSVI_02345_02960 to be in the first column of the name file indicating it is unique. Is this the case?

3

Thank you very much for your reply. I apologize for my delay, I was expecting a notification from the forum (sorry, I did not enable it…) or an answer through github.

Unexpectedly, the WYSVI_02345_02960 problematic label had disappeared from the name file, as you suspected. So, I changed the name and group files to a previous version (I mean ‘done before’ in terms of the pipeline http://www.mothur.org/wiki/Ion_Torrent_sequence_analysis_using_Mothur). The problem remains, but with a different label that, in this case, it is indeed in the first column of the name file:

mothur > summary.tax(taxonomy=run2.shhh.trim.unique.good.filter.nr_v119.wang.taxonomy, name=run2.shhh.names, group=run2.shhh.groups)

reftaxonomy is not required, but if given will keep the rankIDs in the summary file static.
[ERROR]: WYSVI_00540_01863 is not in your name file please correct.
$ egrep WYSVI_00540_01863 run2.shhh.names 
WYSVI_00540_01863 WYSVI_00540_01863

Thanks!

4

Could you send your files to mothur.bugs@gmail.com?

5

Done!
Thanks

6

You have sequences in your taxonomy file with the same name. For example:

line 67484
WYSVI_00540_01863 Bacteria(81);

line 60545
WYSVI_00540_01863 Bacteria(83);

Mothur removes sequences it read in the names file from memory as it reads them in the taxonomy file to save time and space. Mothur expects sequence names to be unique. Did you modify these files outside of mothur?

7

Thank you very much for your help.

No, I did not modify the files outside of mothur. I used MPI against 64 cores, could it be a trouble of wrong data-chunk distribution among cores?