Abstract
COVID-19 often manifests with different outcomes in different patients, highlighting the complexity of the host-pathogen interactions involved in manifestations of the disease at the molecular and cellular levels. In this paper, we propose a set of postulates and a framework for systematically understanding complex molecular host-pathogen interaction networks. Specifically, we first propose four host-pathogen interaction (HPI) postulates as the basis for understanding molecular and cellular host-pathogen interactions and their relations to disease outcomes. These four postulates cover the evolutionary dispositions involved in HPIs, the dynamic nature of HPI outcomes, roles that HPI components may occupy leading to such outcomes, and HPI checkpoints that are critical for specific disease outcomes. Based on these postulates, an HPI Postulate and Ontology (HPIPO) framework is proposed to apply interoperable ontologies to systematically model and represent various granular details and knowledge within the scope of the HPI postulates, in a way that will support AI-ready data standardization, sharing, integration, and analysis. As a demonstration, the HPI postulates and the HPIPO framework were applied to study COVID-19 with the Coronavirus Infectious Disease Ontology (CIDO), leading to a novel approach to rational design of drug/vaccine cocktails aimed at interrupting processes occurring at critical host-coronavirus interaction checkpoints. Furthermore, the host-coronavirus protein-protein interactions (PPIs) relevant to COVID-19 were predicted and evaluated based on prior knowledge of curated PPIs and domain-domain interactions, and how such studies can be further explored with the HPI postulates and the HPIPO framework is discussed.
Introduction
Why do two people infected with the same virus experience vastly different clinical outcomes? Traditional research often focuses in isolation on either the pathogen or the host, but rarely on the complex, dynamic interaction between them at a granular molecular level. Existing models, such as the "Damage Response Framework," provide a high-level view but lack the resolution needed to identify specific causal mechanisms. Furthermore, the explosion of biomedical data has created a "Tower of Babel" problem: data is often stored in heterogeneous, non-interoperable formats that make it nearly impossible for AI and machine learning tools to integrate and analyze effectively.
The HPIPO Framework
To address these challenges, researchers have introduced the HPIPO (Host-Pathogen Interaction Postulate and Ontology) framework. This system establishes a conceptual foundation based on four fundamental postulates:
- Evolutionary Dispositions (P1): Both host and pathogen have evolved "strategies" (e.g., the virus to replicate, the host to destroy the invader).
- Dynamic Outcomes (P2): Disease results from the mechanical interaction of these strategies at the cellular level.
- HPI Roles (P3): Specific molecules play distinct roles, such as a viral protein acting as a "ligand" vs. a host protein acting as a "receptor."
- HPI Checkpoints (P4): Critical "bottlenecks" in the infection process that, if interrupted, can change the disease outcome.
The HPIPO framework translates these postulates into a machine-readable format using ontologies like the Coronavirus Infectious Disease Ontology (CIDO). This allows researchers to use SPARQL queries to find hidden connections between drugs, proteins, and biological processes.
Strategic Insights for COVID-19
A key shift in this framework is moving from measuring general "damage" to identifying specific "checkpoints." These molecular events—such as viral entry, genome replication, or viral release—serve as ideal targets for intervention.
Because infection is a multi-stage process, the framework supports a "cocktail" strategy. Rather than a single "silver bullet," the most effective treatment may be a rational combination of drugs and vaccines designed to hit multiple checkpoints simultaneously. By using the Basic Formal Ontology (BFO) as a top-level structure, the framework ensures that data is FAIR (Findable, Accessible, Interoperable, and Reusable), making it immediately usable for automated AI reasoning.
Key Results
The application of HPIPO to COVID-19 yielded significant findings:
- Drug Discovery: Using CIDO-based SPARQL queries, the team identified 232 drugs with the potential to interrupt COVID-19 checkpoints. Candidates like copper and artenimol were highlighted for their ability to target multiple processes.
- PPI Predictions: The framework predicted 1,001 human-coronavirus protein-protein interactions (PPIs) involving 27 viral proteins and 233 human proteins.
- Validation: When compared against experimentally validated data, the predicted interactions showed a high degree of overlap in functional "Gene Ontology" terms, suggesting high model accuracy.
Conclusion
While powerful, the HPIPO framework's success depends on the continuous curation of knowledge by the scientific community. Moving from correlation to proven molecular causality remains a primary challenge. Future work involves expanding the framework to other emerging infectious diseases and integrating it into automated drug-repurposing pipelines to rapidly identify critical checkpoints for new pathogens.
Summary updated