Disease Resource Information
Defining Disease-Gene Associations
About GUDMAP Disease
The GUDMAP Disease Database uses disease terms obtained from OMIM. In total 8987 OMIM Disease terms are contained within the database.
Method - determining disease/gene associations
Disease terms are obtained by querying OMIM for a range of genitourinary related terms (for example - genitourinary, GU, renal, nephron, urologic etc." within the Clinical Synopsis, Title and Text fields of the OMIM entry. The search is restricted to entries with either the #(pound), %(percent) or 'none' prefix. These three properties ensure that only entries that describe phenotypes are returned (avoiding gene entries).
Disease associations are determined for all genes that have in situ data in GUDMAP. In order to do this an additional text search for the orthologous human gene symbol within the main text of the OMIM entry is done. If the gene symbol is present an association between the disease and the gene is assumed. Associations are then assessed manually after including disease-gene associations from NCBI.
NCBI evidence for gene-disease associations
In addition to text-mining OMIM entries for evidence of disease-gene associations, associations are also obtained direct from NCBI via the mim2genes file - downloaded from the NCBI ftp site (ftp://ftp.ncbi.nlm.nih.gov/gene/DATA/). The mim2genes file is curated by NCBI and is a list of Entrez Gene IDs and OMIM IDs that are associated with one another.
As a consequence of using a text-mining approach some gene symbols, such as 'T' and 'Rest' produces a large number of associations with OMIM entries, the majority of which are likely to be spurious. At present the associations for these types of genes are being checked manually and are not available via the search tools.
Also note that some associations between genes and diseases may not necessarily be positive - e.g. a gene and a disease could be associated because it is known that the gene does not play any role in disease (e.g. Adams-Oliver Syndrome and Msx1). Although the manual checking step should identify these negative associations, it may not always be the case. It is important therefore to follow the link to OMIM to view a more in depth description of the association.
About phenotype associations
The phenotype-gene associations are obtained from the MGI (Mouse Genome Informatics). A gene is considered associated with a particular Mammalian Phenotype (MP) term if a mutant allele of that gene has been annotated with that MP term.
When search for genes associated with a phenotpye term: If a gene has a derived annotation to the MP search term it means that the gene is directly annotated to a different MP term that, according to the MGI MP Ontology, is a child term of the MP search term.
For example, the gene Acvr2a has a derived annotation to abnormal corpus luteum due to the fact that Acvr2a is directly annotated to absent corpus luteum – which, according to the MGI MP Ontology is a child term of abnormal corpus luteum.
When searching for phenotpye terms associated with a gene: Where an MP term has a derived annotation to the gene it is because the gene is directly annotated with a different MP term which, according to the MGI MP Ontology is a child of the other MP term.
For example, abnormal gametogenesis has derived annotation to Wnt4 because Wnt4 is directly annotated with abnormal oogenesis, which, according to the MGI MP Ontology is child term of abnormal gametogenesis.