Vaccine Ontology (VO) and its application in literature mining of vaccine-gene interaction networks
The Vaccine Ontology (VO, http://www.violinet.org/vaccineontology) is a community-based biomedical ontology in the domain of vaccine and vaccination. As of April 29, VO contains 5,578 ontology terms including 5,145 VO-specific terms, and the other terms are imported from over ten existing ontologies. VO represents over 1,000 licensed vaccines and vaccine candidates in clinical trials and in research. These vaccines have been developed against infections of various bacterial, viral, and parasitic pathogens as well as cancers in over 20 animal species. VO also represents all types of vaccine components (e.g., vaccine antigens and adjuvants) and vaccine preparations. In addition, various vaccine-induced immune responses and protection against specific disorders are being modeled in VO. We have developed VO-based natural language processing (NLP) approaches to retrieve and analyze vaccine-specific interaction networks of host or pathogen genes. Our use case studies demonstrated: (1) VO and centrality-based literature mining dramatically increased the identification of interferon-gamma (IFNG) and vaccine-associated human genes and gene interactions; (2) a NLP literature mining method based on VO and the Interaction Network Ontology (INO) allows the detection of significantly over- and under-represented gene-gene interactions in the IFNG network and IFNG-vaccine subnetwork; and (3) VO-based literature mining provided better performance than the MeSH-based PubMed approach in retrieving the interactions between vaccines and genes from Brucella (a bacterial pathogen). Novel scientific hypotheses have been generated through these ontology-based literature mining studies.
Yongqun “Oliver” He is an associate professor in the Unit for Laboratory Animal Medicine, Department of Microbiology, Center for Computational Medicine and Bioinformatics, and Comprehensive Cancer Center of the University of Michigan Medical School, Ann Arbor, MI, USA. He has initiated and led the development of several biomedical ontologies, including the community-based Vaccine Ontology (VO; http://www.violinet.org/vaccineontology) and Ontology of Adverse Events (OAE; http://www.oae-ontology.org). His group, through collaborations with others, has developed several ontology-oriented software programs including OntoFox (http://ontofox.hegroup.org) for importing and reusing individual ontology terms and Ontobee (http://www.ontobee.org) as a linked RDF data server for ontology term information retrieval and display. His group has also developed and applied ontology-based literature mining tools to study biological pathways with an emphasis on vaccine-associated gene interaction networks among host and pathogen genes.