Ontology: The Translational Bridge Between Bioethics and Data-Driven Science
How formal systems for representing knowledge create a common language between ethical reasoning and empirical data
Introduction: The Data-Ethics Tension in Modern Medicine
Imagine a team of medical researchers analyzing thousands of COVID-19 patient records to determine which factors predict severe illness. They identify several genetic markers and pre-existing conditions strongly correlated with hospitalization risk. Immediately, urgent ethical questions emerge: How should this predictive information be used? Could it inadvertently disadvantage certain populations? What obligations do researchers have to act on these findings?
Data-Driven Discovery
The unprecedented scale of biological data generation creates new opportunities for medical insights but also new ethical challenges.
Ethical Reasoning
As we generate more data, the gap between what we can measure and what we should do with that knowledge widens.
Enter ontology—not just as an abstract philosophical concept, but as a practical framework that can bridge the worlds of empirical data and human values. This article explores how formal systems for representing knowledge are creating a common language between bioethics and data science, enabling more nuanced, transparent, and ethically-grounded biomedical research.
What Exactly Is Ontology? From Philosophy to Data Science
Philosophical Foundation
To understand ontology's role in modern bioethics and data science, we must first unpack its dual identity. In philosophy, ontology is the study of being, existence, and reality's fundamental structure 9 . Philosophers working in ontology examine questions like: What types of entities truly exist? How do abstract concepts differ from physical objects? What are the essential properties that define something?
This philosophical tradition provides the conceptual groundwork for understanding how we categorize and make sense of the world 6 . When bioethicists debate whether an embryo has the same moral status as a person, or whether artificial intelligence systems can truly be said to have "agency," they are engaging in ontological inquiry at this philosophical level.
Information Science Revolution
In information science, ontology takes on a more practical meaning. Here, an ontology is a formal representation of knowledge within a domain, consisting of standardized definitions of concepts (classes), their properties, and the relationships between them 8 9 .
These computational ontologies create shared vocabularies that enable both humans and computers to communicate meaning clearly and consistently. As one ontology developer explains, "Not only does this aid the categorization of terms; an ontology encapsulates a broad spectrum of vocabulary within that particular knowledge domain, allowing scientists to better understand the connection between concepts" 8 .
Philosophical vs. Information Science Ontology
| Aspect | Philosophical Ontology | Information Science Ontology |
|---|---|---|
| Primary Focus | Nature of being and reality | Standardized knowledge representation |
| Methods | Conceptual analysis, logical reasoning | Formal logic, computer algorithms |
| Output | Theories about existence | Structured vocabularies, databases |
| Application | Understanding fundamental categories | Data integration, semantic interoperability |
How Ontology Bridges Bioethics and Data Science
The connection between ontology's philosophical and computational forms provides the crucial bridge between bioethics and data-driven inquiry.
Philosophical Understanding
Philosophical ontology helps us understand the nature of ethical concepts
Computational Representation
Computational ontology gives us tools to represent and reason with these concepts systematically
Translational Bridge
This dual capability addresses the fundamental challenge of connecting abstract ethical principles with concrete biomedical contexts 5
The Power of Formalization
Computational ontologies like EXPO (an ontology of scientific experiments) formalize generic concepts of experimental design, methodology, and results representation 1 . When a clinical trial is described using such formal structures, the ethical dimensions—informed consent procedures, risk-benefit assessments, data handling protocols—become explicit, searchable, and analyzable rather than buried in unstructured text.
Consider the Gene Ontology (GO), one of the most successful biomedical ontologies, which provides a standardized framework for describing gene functions across three categories: biological processes, molecular functions, and cellular components 7 .
GO allows researchers to systematically analyze which biological pathways are affected in a particular disease state. But it also enables more nuanced ethical analysis: by making functional relationships explicit, GO helps researchers identify when a gene involved in a disease process might also play roles in other biological systems, raising important ethical questions about potential side effects of gene-based therapies.
Gene Ontology Framework
GO provides a structured framework for analyzing gene functions across biological processes, molecular functions, and cellular components.
Case Study: An Ontological Approach to Ethical Questions in Breast Cancer Genomics
The Experimental Challenge
A recent investigation into the genomics of breast cancer employed GO analysis to pinpoint significantly enriched terms related to cell proliferation, apoptosis, and DNA repair mechanisms 7 . This analysis underscored vital pathways, such as the p53 signaling cascade, offering key insights into tumor progression.
While scientifically valuable, such findings immediately raise ethical questions:
- Should patients be informed of genetic markers with uncertain clinical significance?
- How should researchers handle incidental findings?
- What are the justice implications if resulting treatments are prohibitively expensive?
Genomic research in breast cancer raises important ethical questions about data use and patient communication.
Methodology and Ontological Framework
1. Experimental Formalization
The research followed principles similar to EXPO ontology, making explicit their experimental design, methodology, and results representation 1 .
2. Data Annotation
Genetic data were annotated using Gene Ontology terms, creating structured information that could be ethically analyzed alongside clinical parameters.
3. Enrichment Analysis
Statistical methods identified overrepresented GO terms in breast cancer samples compared to normal tissue.
4. Ethical Mapping
Ethical dimensions were mapped onto the structured data using relationship extraction techniques similar to those described in chemical ontologies 8 .
Breast Cancer Genomics: Key Findings and Ethical Dimensions
| Biological Finding | GO Term | Ethical Dimension | Bioethical Question |
|---|---|---|---|
| p53 pathway disruption | GO:0007156 (cell adhesion) | Justice | Will targeted therapies be accessible across socioeconomic groups? |
| DNA repair mechanisms enriched | GO:0006281 (DNA repair) | Privacy | Should relatives be notified of inherited cancer risks? |
| Cell proliferation signals | GO:0008283 (cell proliferation) | Consent | How to communicate uncertainty in risk assessment? |
Results and Significance
The ontological approach enabled researchers to not only identify biologically significant pathways but also to systematically associate these findings with their ethical implications. The study demonstrated "a significant improvement in the identification of enriched GO terms" 7 , with the structured framework revealing connections that might have been overlooked in traditional bioethical analysis. By using GO analysis, the researchers transformed raw genomic data into biologically meaningful patterns, which could then be subjected to systematic ethical analysis rather than ad hoc consideration.
Methodology: The Process of Ontology-Assisted Bioethics Research
Planning & Requirements
Researchers identify key ethical concepts and relationships relevant to their domain, determining what needs to be represented in the ontology.
Design
Formalizing concepts using standardized ontological relationships such as "is_a" (for categorization) and "part_of" (for mereological relationships) .
Validation & Verification
Ensuring the ontological framework adequately represents the ethical landscape through expert review and use case testing.
Implementation
Using the framework to analyze specific cases or datasets, integrating with existing research workflows and data systems.
The Scientist's Toolkit: Ontological Resources for Bioethics Research
Researchers exploring the intersection of bioethics and data science can leverage several ontological resources:
| Resource | Type | Application in Bioethics | Access |
|---|---|---|---|
| EXPO Ontology | General experiment ontology | Formalizing ethical aspects of study design 1 | Public |
| Gene Ontology (GO) | Gene function ontology | Identifying ethical implications of functional pathways 3 7 | Public |
| OC Processor | Relationship extraction tool | Mapping connections between data and ethical concepts 8 | Commercial |
| SUMO | Upper ontology | Providing foundational concepts for ethical frameworks 1 | Public |
Implementing Ontological Analysis
The process of applying ontological methods to bioethics research typically involves several stages, mirroring the ontology development process described by OntoChem: planning and requirements, design, validation and verification, and implementation 8 .
In the planning phase, researchers identify the key ethical concepts and relationships relevant to their domain. During design, they formalize these concepts using standardized ontological relationships.
Validation & Implementation
Validation ensures the ontological framework adequately represents the ethical landscape, while implementation involves using the framework to analyze specific cases or datasets.
This structured approach helps ensure that ethical considerations are systematically integrated into data-driven research rather than being treated as an afterthought.
Future Directions: Where Ontology Could Take Bioethics
The integration of ontological methods into bioethics is still emerging, but several promising directions are appearing. As one researcher notes, ontology serves as "a bridge between bioethics and data-driven inquiry" 5 , suggesting several future applications:
Predictive Ethics
By formally representing ethical frameworks and past cases, ontologies could help identify potential ethical issues in study designs before implementation.
Transparency and Explanation
Ontological representations make ethical reasoning more explicit and transparent, potentially helping patients and research subjects understand how decisions are made.
Cross-Cultural Bioethics
Formal ontologies could help identify where ethical principles are universal versus culturally specific, though this requires careful attention to power dynamics in who defines the ontologies.
Regulatory Science
Ontologies could streamline ethics review processes by creating structured representations of research protocols and their ethical dimensions.
Technical Challenges and Opportunities
While promising, the integration of ontology into bioethics faces several challenges:
- Complexity of Ethical Concepts: Many ethical concepts resist simple formalization
- Interdisciplinary Collaboration: Requires close work between ethicists, domain experts, and computer scientists
- Dynamic Nature of Ethics: Ethical frameworks evolve over time, requiring adaptable ontological systems
- Implementation Barriers: Integrating ontological approaches into existing research workflows
Conclusion: Toward an Ethically Grounded Data Science
Ontology, in both its philosophical and computational senses, provides the translational bridge that allows bioethics and data science to inform each other in meaningful ways.
Philosophical Foundation
The philosophical tradition of ontology gives us tools to understand the nature of ethical concepts, providing the conceptual groundwork for ethical analysis.
Computational Implementation
Computational ontology provides methods to represent and reason with these concepts systematically, enabling integration with data-driven research.
As we've seen through examples like the Gene Ontology and EXPO, formal knowledge representations make both scientific and ethical assumptions explicit, enabling more transparent and accountable research practices. The breast cancer case study illustrates how ontological thinking can reveal connections between molecular pathways and ethical considerations that might otherwise remain separate domains of inquiry. As biomedical research becomes increasingly data-driven, the integration of ontological methods offers a promising path toward maintaining ethical integrity while embracing the power of data science. Ontology doesn't replace ethical judgment, but it provides the structural framework that makes such judgment more systematic, transparent, and grounded in both data and values.
The future of bioethics may depend on building better bridges between human values and data—and ontology provides the architectural blueprint for these essential connections.