.. _pick_closed_reference_otus:

.. index:: pick_closed_reference_otus.py

*pick_closed_reference_otus.py* -- Closed-reference OTU picking/Shotgun UniFrac workflow.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

**Description:**

This script picks OTUs using a closed reference and constructs an OTU table. Taxonomy is assigned using a pre-defined taxonomy map of reference sequence OTU to taxonomy. If full-length genomes are provided as the reference sequences, this script applies the Shotgun UniFrac method.


**Usage:** :file:`pick_closed_reference_otus.py [options]`

**Input Arguments:**

.. note::

	
	**[REQUIRED]**
		
	-i, `-`-input_fp
		The input sequences
	-r, `-`-reference_fp
		The reference sequences
	-o, `-`-output_dir
		The output directory
	
	**[OPTIONAL]**
		
	-p, `-`-parameter_fp
		Path to the parameter file, which specifies changes to the default behavior. See http://www.qiime.org/documentation/file_formats.html#qiime-parameters . [if omitted, default values will be used]
	-t, `-`-taxonomy_fp
		The taxonomy map [default: None]
	-f, `-`-force
		Force overwrite of existing output directory (note: existing files in output_dir will not be removed) [default: None]
	-w, `-`-print_only
		Print the commands but don't call them -- useful for debugging [default: False]
	-a, `-`-parallel
		Run in parallel where available [default: False]
	-O, `-`-jobs_to_start
		Number of jobs to start. NOTE: you must also pass -a to run in parallel, this defines the number of jobs to be started if and only if -a is passed [default: 2]


**Output:**




Pick OTUs, assign taxonomy, and create an OTU table against a reference set of OTUs. ALWAYS SPECIFY ABSOLUTE FILE PATHS (absolute path represented here as $PWD, but will generally look something like /home/ubuntu/my_analysis/).

::

	pick_closed_reference_otus.py -i $PWD/seqs.fna -r $PWD/refseqs.fna -o $PWD/otus_w_tax/ -t $PWD/taxa.txt

Pick OTUs and create an OTU table against a reference set of OTUs without adding taxonomy assignments. ALWAYS SPECIFY ABSOLUTE FILE PATHS (absolute path represented here as $PWD, but will generally look something like /home/ubuntu/my_analysis/).

::

	pick_closed_reference_otus.py -i $PWD/seqs.fna -r $PWD/refseqs.fna -o $PWD/otus/

Pick OTUs, assign taxonomy, and create an OTU table against a reference set of OTUs using usearch_ref. ALWAYS SPECIFY ABSOLUTE FILE PATHS (absolute path represented here as $PWD, but will generally look something like /home/ubuntu/my_analysis/).

::

	pick_closed_reference_otus.py -i $PWD/seqs.fna -r $PWD/refseqs.fna -o $PWD/otus_usearch/ -p $PWD/usearch_params.txt -t $PWD/taxa.txt


