Knowledge Graph and Machine Learning Help the Research of Drugs Aimed at Neurological Diseases

This study transcends conventional network models by embracing the vast semantic diversity of relationships among biomedical entities. Gathering vast datasets from diverse sources, spanning numerous entity types, we have integrated them to create the comprehensive knowledge graph named PATHOS (PATHologies of HOmo Sapiens). PATHOS serves as the foundation for our knowledge-driven predictions. Subsequently, we implemented LOGOS (Learning Optimized Graph-based representations of Object Semantics), a knowledge graph embedding model capable of generating predictions relevant to drug research.
The choice of the name PATHOS pays homage to ancient Greek culture. In this context, "pathos" ($\pi \acute{\alpha} \theta o \varsigma$) relates to emotions, sufferings and experiences. In the philosophical tradition, "logos" ($\lambda \acute{o} \gamma o \varsigma$) represents rationality and explanation. This aligns seamlessly with the knowledge graph embedding model's function of providing reasoned predictions based on the structured knowledge contained within the PATHOS knowledge graph. In this pairing, LOGOS complements PATHOS by offering rational insight into the depths of human sufferings.
PATHOS and LOGOS demonstrated their potential in three paradigmatic case studies (with a focus on neurological diseases): drug repurposing for Alzheimer's disease, phenotype selection for Huntington's disease, and the identification of proteins linked to multiple sclerosis.

License Disclamer

PATHOS and LOGOS are made available under the MIT license. If you intend to incorporate them into other works or use it in a publication or your research, please cite us.
PATHOS contains resources which incorporate material contributed or licensed by individuals, companies, or organizations that may be protected by copyright laws. All persons reproducing, redistributing, or making commercial use of this information are expected to adhere to the terms and conditions asserted by the copyright holder.
A detailed list of the licenses of all sources is available in the sources.json file.