Automated ontological gene annotation for computing disease similarity.
Year: 2010
Type: article
Abstract: The annotation of gene/gene products with information on associated diseases is useful as an aid to clinical diagnosis and drug discovery. Several supervised and unsupervised methods exist that automate the association of genes with diseases, but relatively little work has been done to map protein sequence data to disease terminologies. This paper augments an existing open-disease terminology, the Disease Ontology (DO), and uses it for automated annotation of Swissprot records. In addition to the inherent benefits of mapping data to a rich ontology, we demonstrate a gain of 36.1% in gene-disease associations compared to that in DO. Further, we measure disease similarity by exploiting the co-occurrence of annotation among proteins and the hierarchical structure of DO. This makes it possible to find related diseases or signs, with the potential to find previously unknown relationships.
Source: PubMed
Authors: Sachin Mathur, Deendayal Dinakarpandian
Institution: University of Missouri–Kansas City
Language: English
Subfield: Health Informatics
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W341993957