GO subsets

What are GO subsets?

GO subsets (also known as GO slims) are simplified versions of the GO containing a reduced number of terms. There are several types of subsets:

Binning: Binning subsets are intended to be used to summarize the molecular function/biological processes/cellular components for sets of genes, including entire genomes/proteomes.
Ribbon: Ribbons are used to give a quick overview of the broad classes of terms annotated for a gene. Ribbon subsets are usually smaller than binning subsets.
Exclusion List: Exclusion lists are terms that should not be used for annotations, either because the term is too broad to be informative (for example, “GO:0008152 metabolic process”), and/or because a more specific term should be used (for example, more specific children of “GO:0048856 anatomical structure development” should be used that describe the actual structure formed).

How are GO subsets used?

GO subsets are particularly useful for providing an overview of the range of functions and processes found in a given clade or organism’s genome. Given a coarse grained view of the ontology content without the detail of the specific fine-grained terms, these slims can offer an overall sense of the key biological functions that are vital to an organism. For example, the limited number of opsin genes in bedbugs, or the abundance of kinins in ticks.
GO subsets are also useful for addressing specific research needs in particular areas of biology. For instance, if a researcher is specifically interested in the process of development or in the nuclear proteome, subsets allow them to focus solely on the terms under that section of the ontology.
GO subsets can also be useful for simplifying searches or annotation operations. By reducing the number of choices a user is presented with their overall efficiency can be improved.

Who creates GO subset?

GO provides a generic subset which, like the GO itself, is species-agnostic, and which should be suitable for most purposes. In addition, many model organism-specific subsets have been created by GO consortium members and are available for download as listed below.
Users and user communities can create their own GO subsets. Please contact the GO helpdesk for more information about creating and submitting your GO subsets.

How are GO subsets maintained?

Groups who have created a GO subset are responsible for keeping it up to date as the ontology changes. GO subsets are periodically reviewed to make sure they are relevant, that pertinent new GO terms are added, and that obsolete GO terms are replaced by valid terms to ensure the appropriate grouping of annotations.

Download GO subsets

GO subsets are part of the ontology, under the tag subset.
For example: [Term] id: GO:0048856 name: anatomical structure development namespace: biological_process subset: gocheck_do_not_annotate subset: goslim_chembl subset: goslim_generic
GO subsets can also be downloaded as separated files in the download ontology section.
The files available below for download are generated by script from that file.
GO subsets (slims) are available in OBO. OWL as as well as JSON; JSON files should be loaded with ontobio, although they can be opened with any text editor.

Subset name	Maintainer	File name	OBO format	OWL format	json format	tsv format
Alliance of Genome Resources subset	Developed by GO Consortium for the Alliance of Genomes Resources	goslim_agr	obo	owl	json	tsv
Generic GO subset	GO Consortium	goslim_generic	obo	owl	json	tsv
Aspergillus subset	Aspergillus Genome Data	goslim_aspergillus	obo	owl	json	tsv
Candida albicans subset	Candida Genome Database	goslim_candida	obo	owl	json	tsv
Drosophila subset	FlyBase	goslim_drosophila	obo	owl	json	tsv
Chembl Drug Target subset	ChEMBL	goslim_chembl	obo	owl	json	tsv
Metagenomics subset	InterPro group	goslim_metagenomic	obo	owl	json	tsv
Mouse GO subset	Mouse Genome Informatics	goslim_mouse	obo	owl	json	tsv
Plant subset	The Arabidopsis Information Resource	goslim_plant	obo	owl	json	tsv
Prokaryote subset	GO Consortium	goslim_prokaryote	obo	owl	json	tsv
Protein Information Resource subset	PIR	goslim_pir	obo	owl	json	tsv
Schizosaccharomyces pombe subset	PomBase	goslim_pombe	obo	owl	json	tsv
Yeast subset	Saccharomyces Genome Database	goslim_yeast	obo	owl	json	tsv

GO “anti-slims”

For internal checking purposes, GO maintains two “anti-slims”, terms to which annotations should not be made. “Anti-slim” terms should never be used when creating a subset, and terms that are obsoleted are removed from subsets.

Subset name	Usage	File name	OBO format	OWL format	json format	tsv format*
Do not annotate	The set of high level terms that are useful for grouping, but should have no direct annotations	gocheck_do_not_annotate	obo	owl	json	tsv
Obsoletion candidate	GO terms planned for obsoletion. This subsets serves as an early warning system both for users and curators	gocheck_obsoletion_candidate	obo	owl	json	tsv

Map2Slim option in OWLTools: Given a GO subset, a current ontology, and specific gene product associations to precise terms from this current ontology, the Map2Slim script will map these term associations up to their respective broader terms from the GO subset. This script is an option of OWLTools, and it can be used to either create a new gene associations file, which contains the same annotations but made to the subset terms; or in count-mode, in which case it will give distinct gene product counts for each subset term. Background information and details on how to download, install, and implement OWLTools, as well as instructions on how to run the Map2Slim script are available from the OWLTools Wiki.
On the web: Similarly, there are online tools that may be of use including the Princeton slimming tool. It should be noted that online tools do often contain limitations and timeouts.