NAME

go-slimdown.pl

SYNOPSIS

go-slimdown.pl -i go/ontology/gene_ontology.obo -s goslim_generic -o slimmed.obo

DESCRIPTION

# must supply these arguments... or else! # INPUT -i || --ontology /path/to/<file_name> input file (ontology file) with path

# OUTPUT -o || --output /path/to/<file_name> output file with path or -b || --basename /path/to/<file_name_containing_SLIM_NAME>

specify a file name containing the text "SLIM_NAME", which will be
substituted with the name of the subset
e.g. -s goslim_goa -s goslim_yeast -b /temp/gene_ontology.SLIM_NAME.obo
would produce two files,
/temp/gene_ontology.goslim_goa.obo and /temp/gene_ontology.goslim_yeast.obo

# SUBSET -s || --subset <subset_name> name of the subset to extract; multiple subsets can be specified or -a || --get_all_subsets extract all the subsets in the graph

# optional args -c || --combined if more than one subset is specified, create a slim using terms from all of the subsets specified

 -v || --verbose                         prints various messages

	Given a file where certain terms are specified as being in subset S, this
	script will 'slim down' the file by removing terms not in the subset.

	Relationships between remaining terms are calculated by the inference engine.

	If the root nodes are not already in the subset, they are added to the graph.

	The slimming algorithm is 'relationship-aware', and finds the closest node
	for each relation (rather than just the closest term). For example, if we
	had the following relationships:

	A -i- B -i- C -p- D

	the slimmer would say that B was the closest node via the 'i' relation, and
	D was the closest via the 'p' relation.

	Note that there may be several different relationships between the same two
	terms in the slimmed file.