categorize_by_function.py – Collapse table data to a specified level in a hierarchy.

Description:

This script collapses hierarchical data to a specified level. For instance, often it is useful to examine KEGG results from a higher level within the pathway hierarchy. Many genes are sometimes involved in multiple pathways, and in these circumstances (also know as a one-to-many relationship), the gene is counted for each pathway. This has a side effect of increasing the total count of genes in the table.

Usage: categorize_by_function.py [options]

Input Arguments:

Note

[REQUIRED]

-i, --input_fp

The predicted metagenome table

-o, --output_fp

The resulting table

-c, --metadata_category

The metadata category that describes the hierarchy

-l, --level

The level in the hierarchy to collapse to. A value of 0 is not allowed, a value of 1 is the highest level, and any higher value nears the leaves of the hierarchy. For instance, if the hierarchy contains 4 levels, specifying 3 would collapse at one level above being fully specified.

[OPTIONAL]

--ignore

Ignore the comma separated list of names. For instance, specifying –ignore_unknown=unknown,unclassified will ignore those labels while collapsing. The default is to not ignore anything. [default: None]

-f, --format_tab_delimited

Output the predicted metagenome table in tab-delimited format [default: False]

Output:

Output table is contains gene counts at a higher level within a hierarchy.

Collapse predicted metagenome using KEGG Pathway metadata.

categorize_by_function.py -i predicted_metagenomes.biom -c KEGG_Pathways -l 3 -o predicted_metagenomes.L3.biom

Change output to tab-delimited format (instead of BIOM).

categorize_by_function.py -f -i predicted_metagenomes.biom -c KEGG_Pathways -l 3 -o predicted_metagenomes.L3.txt

Collapse predicted metagenome using taxonomy metadata (not one-to-many).

categorize_by_function.py -i observation_table.biom -c taxonomy -l 1 -o observation_table.L1.biom