The RGO is released as a MySQL dump consisting in the following five tables:
- Terms contains data from the original GO (name, term_type, acc, is_root, is_relation). Each term contains also a reference (the node_id) to the RGO node that contains it.
- Synonyms contains a subset of the Synonym table of the GO.
- Nodes describes the RGO nodes. Attributes are the id (referenced by Terms.node_id) and the node_type (one of the three sub-ontologies).
- Edges contains the linking information. Relevant attributes are weight, and native. Attribute native is a boolean set to be 0 if the edge is a cross-ontology edge (i.e., if the source and the destination nodes do not belong to the same sub-ontology); it is set to 1 otherwise.
- Annotations: contains manually curated annotations for Homo Sapiens and Saccharomyces Cerevisiae. Attributes are:
- symbol the standard name of the annotated gene/protein;
- taxa the TAXA id for the specie under study (9606 for Human, 4932 for Yeast);
- term_id the term identifier upon which the gene/protein is annotated;
- evidence code quality of annotation. They are the same of GO.
Untar and unzip the file (on unices: tar -xzvf RGO_201003.tar.gz)
Create a database for the RGO (on unices: mysqladmin [options] create rgo)
Import the dump file in the database (on unices: cat RGO201003.sql | mysql [options] -D rgo)