Regarding OTUS assigned to a particular taxa

[Muneer Malla]

 Hello All

 I am facing a problem 

I want to calculate the number of OTUS that are assigned to a particular taxa.

Please suggest me the command.

[TonyWalters]

Hello,

Perhaps inelegant but this should work:

convert your .biom table to a tab delimited form, see:

http://biom-format.org/documentation/biom_conversion.html

specifically this example:

biom convert -i otu_table.biom -o otu_table.txt --to-tsv --header-key taxonomy

Then open your table in Excel, sort by taxonomy (right-most column) and then find the taxa in question and see how many rows there are with it.

Alternatively, you could use the .txt convertd file and do:

grep -c "X" otu_table.txt

where X is the taxonomy in question.

[Muneer Malla]

Thanks Tony 

I tried it and it worked. 

However for clarification I am attaching the file

While I looked for Proteobacteria  and I calculate it in rows and it sums up 46.

Second Point now is how can I calculate the mean relative abundance of this group. along with others groups as well. 

[Marco Severgnini]

Hello Muneer, if you want to calculate the relative abundance or the absolute count of reads in all your taxa, the best way should be using the "summarize_taxa.py" script.

The syntax should be the following:

summarize_taxa.py -i otu_table.biom -o tax_mapping/ [-L 2,3,4,5,6] [-a]

• The "-L" option allows you to specify the taxonomic level of interest. If you are interested only to "Proteobacteria" and other phyla, you could specify "-L 2". Otherwise, by default, QIIME calculates the relative abundances of all levels between phylum (L2) and genera (L6).
• The "-a" option, on the other hand, allows you to have a table with absolute counts instead of fractions.

You can see all the options with:

summarize_taxa.py -h

Marco

[Muneer Malla]

Thanks Marco
The way Tony last mentioned it I have done that way and I think it worked and I got the exact no of OTUs for each taxa.
However I will also try to run this command.

[TonyWalters]

Hello,

Try this:

biom normalize-table --relative-abund -i input_table -o output_table

using the table created earlier via the biom convert comment as the input_table. This should create a relative abundance form of the table that you can see the individual OTUs in (rather than collapsed by taxa as in the summarize_taxa.py output). 

[Muneer Malla]

Hello Tony
Well I tried the command biom normalize-table --relative-abund
However the command gave the following error
Traceback (most recent call last):
  File "/usr/local/bin/pyqi", line 184, in <module>
    optparse_main(cmd_obj, argv[1:])
  File "/usr/local/lib/python2.7/dist-packages/pyqi/core/interfaces/optparse/__init__.py", line 275, in optparse_main
    result = optparse_cmd(local_argv[1:])
  File "/usr/local/lib/python2.7/dist-packages/pyqi/core/interface.py", line 41, in __call__
    return self._output_handler(cmd_result)
  File "/usr/local/lib/python2.7/dist-packages/pyqi/core/interfaces/optparse/__init__.py", line 250, in _output_handler
    opt_value)
  File "/usr/local/lib/python2.7/dist-packages/biom/interfaces/optparse/output_handler.py", line 80, in write_biom_table
    table.to_hdf5(f, generatedby())
  File "/usr/local/lib/python2.7/dist-packages/biom/table.py", line 3484, in to_hdf5
    self.group_metadata(axis='observation'), 'csr', compression)
  File "/usr/local/lib/python2.7/dist-packages/biom/table.py", line 3456, in axis_dump
    formatter[category](grp, category, md, compression)
  File "/usr/local/lib/python2.7/dist-packages/biom/table.py", line 3440, in vlen_list_of_str_formatter
    data[i, :len(value)] = value
TypeError: len() of unsized object
One more thing the file is generated I am also attaching the file

[TonyWalters]

Try your original biom file instead of the .txt file then (NOT the output from summarize_taxa.py).

[Muneer Malla]

Hi again
Sorry I tried the file is generated but upon Opening the file it gives error.
Secondly I have calculate relative abundance in the excel sheet that you mentioned by considering the  OTU of a particular taxa upon the total OTUs .
Will that work or not.

[Muneer Malla]

Hello again 

I am successfully done with the command. After running the 

biom normalize-table --relative-abund -i input_table -o output_table

I generated the table by was unable to open it 

i then ran the command Biom convert and it worked. 
Here I am attaching the table 

Please let me know how I can calculate the overall relative abundance of one taxa across all the three samples.

[Muneer Malla]

Hello Tony 

Are U there 

I am waiting to get my confusion cleared.

[TonyWalters]

This board isn't actively monitored, just so you know-all current development is with QIIME2, not QIIME1

If you want to see the relative abundance of one particular taxa, just use the summarize_taxa.py output, and perhaps increase the level to include 7.

Otherwise, you'll need to do your own calculations-get the sum of counts in an absolute counts table across all taxa, then just divide your target taxa by those total counts to get the relative abundance.