The paper presents a methodology for defining formal, unambiguous relations in biomedical ontologies to improve consistency and interoperability. It addresses the lack of formal definitions for relations in existing ontologies, which can lead to inconsistent usage and errors in annotation. The authors propose a Relation Ontology that provides rigorous definitions for key relations such as is_a, part_of, located_in, and others, ensuring clarity and consistency across ontologies. The methodology involves defining relations in terms of classes and instances, distinguishing between different types of relations (class-class, instance-class, instance-instance), and incorporating temporal and spatial aspects. The paper also discusses the importance of distinguishing between continuants (entities that persist over time) and processes (temporal events), and how these concepts influence the definition of relations. The authors emphasize the need for formal definitions that are both precise and accessible to ontology developers and users, enabling reliable automated reasoning and integration of biological knowledge. The proposed Relation Ontology aims to support the construction of more robust and interoperable ontologies in the life sciences.The paper presents a methodology for defining formal, unambiguous relations in biomedical ontologies to improve consistency and interoperability. It addresses the lack of formal definitions for relations in existing ontologies, which can lead to inconsistent usage and errors in annotation. The authors propose a Relation Ontology that provides rigorous definitions for key relations such as is_a, part_of, located_in, and others, ensuring clarity and consistency across ontologies. The methodology involves defining relations in terms of classes and instances, distinguishing between different types of relations (class-class, instance-class, instance-instance), and incorporating temporal and spatial aspects. The paper also discusses the importance of distinguishing between continuants (entities that persist over time) and processes (temporal events), and how these concepts influence the definition of relations. The authors emphasize the need for formal definitions that are both precise and accessible to ontology developers and users, enabling reliable automated reasoning and integration of biological knowledge. The proposed Relation Ontology aims to support the construction of more robust and interoperable ontologies in the life sciences.