Curating Biomedical Ontologies with Modern AI Techniques

Project description

An ontology is a formal, explicit and shared representation of conceptual knowledge, varying from simple tree-like taxonomies and graph structured semantic networks to complicated modern OWL ontologies underpined by Description Logics (DL). Ontologies have been widely used for representing formal domain knowledge and supporting logical reasoning, in many domains including biomedicine. Quite a few biomedical ontologies, such as SNOMED CT, the Human Disease Ontology (DOID), the Orphanet Rare Disease Ontology (ORDO) and the International Classification of Diseases (ICD), have played a fundamental role in biomedical information systems and AI applications in health science.

High quality biomedical ontologies rely on continuous curation which involves quite a few challenging tasks such as new concept insertion, concept resolution, ontology completion with subsumptions and logical expressions, subontology extraction, content sharing and ontology integration. The current ontology management tools like Prot??g?? still rely on human beings for finishing most of these tasks, and automation remains a big challenge. With the development of AI, many modern machine learning and natural language processing techniques, such as large language model, graph neural network and knowledge graph embedding, have shown great potential to automatically learn and predict over graph structures, literals and logical axioms of ontologies, and thus it becomes a promising direction to curate biomedical ontologies by exploiting these modern AI techniques in combination with knowledge representation and reasoning.

The Department of Computer Science at the University of Manchester invites applications for PhD candidates in the area of biomedical ontology curation. PhD projects in this area will explore how contemporary techniques in Machine Learning, Natural Language Processing and Knowledge Engineering can be used for automatically curating biomedical ontologies.

Applicants are expected to have:

1. An excellent undergraduate degree in Computer Science, Bioinformatics or Mathematics (or related discipline), and preferably, a relevant MSc degree.
2. Confidence and independence in programming complex systems in Java or Python.
3. Previous academic or industry experience in at least one of the relevant topics such as Machine Learning, Natural Language Processing, Bioinformatics, Semantic Web and Knowledge Engineering.
4. Excellent report writing and presentation skills.

Please note that applicants must additionally satisfy the standard requirements for postgraduate studies at the University of Manchester, such as a first-class or high upper-second class (or an equivalent international qualification) and English language qualifications, as stated in the PGR guidelines.