Troubles with Greengenes database 13.8

[metabar75]

Hi all,
I used Greengenes 13.8 to produce the input file for PICRUSt since I read in the forum that Greengenes 13_8 is a minor release and shares an identical OTU structure with 13_5, and so it is okay to use data picked against 
13_8 with PICRUSt. But when I tried to normalize by copy number I got the following error message:

Traceback (most recent call last):
  File "/net/rcstore02/ifs/rc_labs/galaxy/www/galaxy_hutlab14/galaxy-hutlab14-20150109/tools/picrust_galaxy/scripts/normalize_by_copy_number.py", line 146, in <module>
    main()
  File "/net/rcstore02/ifs/rc_labs/galaxy/www/galaxy_hutlab14/galaxy-hutlab14-20150109/tools/picrust_galaxy/scripts/normalize_by_copy_number.py", line 98, in main
    count_table = convert_precalc_to_biom(count_table_fh,ids_to_load)
  File "/usr/local/galaxy-python/lib/python2.7/site-packages/picrust/util.py", line 87, in convert_precalc_to_biom
    raise ValueError,"One or more OTU ids were not found in the precalculated file!\nAre you using the correct --gg_version?\nExample of (the {0}) unknown OTU ids: {1}".format(len(ids_to_load),', '.join(list(ids_to_load)[:5]))
ValueError: One or more OTU ids were not found in the precalculated file!
Are you using the correct --gg_version?
Example of (the 23504) unknown OTU ids: 11540, 11547, 11545, 89378, 5988

Any help will be very much appreciated!
Joan

[Melissa Lumish]

Hi,

 

I am having a similar problem - I actually downloaded Greengenes 13_5 and used this to produce the input file for PICRUSt, but I am continuing to get a similar error message, stating that one or more OTUs were not found.  If anyone is able to help, that would be great!

[Morgan Langille]

Can you send me your otu table directly?

Morgan

--
You received this message because you are subscribed to the Google Groups "picrust-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to picrust-user...@googlegroups.com.
To post to this group, send email to picrus...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Jesse Zaneveld]

Hi,

Thanks for the question.

Could either of you send the command used to generate your OTU table?

Jesse

--

[MLmicrobe]

I have tried to generate the OTU table by modifying the UPARSE pipeline:

Currently, this looks like the following: assign_taxonomy.py -i total.5.repset.fasta -o tax --id_to_taxonomy_fp /Users/Shared/miseq\ pools/uparse_2/gg_13_5_taxonomy --reference_seqs_fp /Users/Shared/miseq\ pools/uparse_2/gg_13_5.fasta --assignment_method mothur --confidence 0

I then tried using the exact closed reference OTU picking strategy provided by the Picrust website and then used the otu_table.biom file found in the ucr97 folder for upload into Picrust. This still produced over 2000 unknown otus.

[MLmicrobe]

Also, does it matter which genome you assign when you upload the file to the online/galaxy version of Picrust? (For example Mouse 2007, human 2009)

[MLmicrobe]

pick_closed_reference_otus.py -i seqs.fna  -o ucr97 -p otu_picking_params_97.txt -r gg_13_5.fasta -t gg_13_5_taxonomy

[MLmicrobe]

Resolution! I was able to get this to work.  I was initially getting an error message related to poor recognition of the biom format, so I converted the file from biom to -json format, which is similar.  Next, although I chose OTUs using closed-reference OTU picking, I was continuing to get the error message that indicated some OTUs were not recognized. Therefore, I used a filter function in order to filter out OTUs that were not recognized by the greengenes database.  I was not able to download the 13_5_otus, so instead I used 13_8_otus, which as was stated in prior threads, should be compatible with PICRUSt.  Please share thoughts regarding this resolution!

Conversion to json format:

biom convert -i otu_table.biom --to-json --table-type "OTU table" -o otu_table_new.biom

Filter otus:

filter_otus_from_otu_table.py -i otu_table_new.biom -o closed_otu_table.biom --negate_ids_to_exclude -e /macqiime/greengenes/gg_13_8_otus/rep_set/97_otus.fasta