Publicação

The ImmunoPeptidomics Ontology : design and evaluation

Ver documento

Detalhes bibliográficos
Resumo:There is significant hardship in manipulating the complex and variable data generated by biomedical subdomains such as immunopeptidomics. Therefore, ontologies are often used to express knowledge in a domain to help establish a standard nomenclature to help handle and integrate medical data. As a new field, immunopeptidomics lacks standardization: there is no recognized vocabulary or explicitly defined meanings. Hundreds of ontologies cover the biomedical domain, including neighboring subdomains like proteomics and immunology, but none adequately cover immunopeptidomics. This shortcoming needed to be addressed so cancer research in this domain could start to affect clinical practice. This dissertation details the development of the ImmunoPeptidomics Ontology, ImPO, to allow later integration of data produced by immunopeptidomics in personalized oncology. For this purpose, ImPO was designed following a process that comprised: capturing domain specialist knowledge in immunopeptidomics; iterative conceptual modeling of the domain through an Entity-Relationship model, semantic modeling by OWL formalization of the ER, cross-referencing ImPO with 28 external ontologies, and evaluation with competency questions and construction pitfalls. Unlike most biomedical ontologies currently accessible, ImPO was created to be populated with data and function as the semantic backbone of a knowledge graph (KG). ImPO was created as part of the KATY project, which aims to apply “AI-empowered knowledge” to clinical practice in Clear Cell Renal Cell Carcinoma. ImPO is one of the KATY KG components that will facilitate data integration in the project and give explainability to the AI techniques created in the project. Nonetheless, the ImPO ontology was created to be utilized independently of the KATY KG as a stand-alone knowledge model to aid in data integration and knowledge discovery in immunopeptidomics.
Autores principais:Eugénio, Patrícia Raquel dos Anjos
Assunto:Ontologia Imunopeptidómica ImPO Teses de mestrado - 2023
Ano:2023
País:Portugal
Tipo de documento:dissertação de mestrado
Tipo de acesso:acesso aberto
Instituição associada:Universidade de Lisboa
Idioma:inglês
Origem:Repositório da Universidade de Lisboa

Registos relacionados

Registos relacionados