Extending inherited metabolic disorder diagnostics with biomarker interaction visualizations.

Area of Science:

• Biochemistry
• Medical Genetics
• Computational Biology

Background:

• Inherited Metabolic Disorders (IMDs) present diagnostic challenges due to non-specific symptoms and complex biomarker interactions.
• Existing diagnostic methods struggle with overlapping biomarkers and lack of clear genotype-phenotype correlations in IMDs.
• Visualizing metabolic interactions could significantly aid in the diagnosis of these rare diseases.

Purpose of the Study:

• To develop and test a proof-of-concept framework for integrating metabolic pathway knowledge with patient data for diagnosing IMDs.
• To provide a visualization tool that connects metabolic biomarkers and enzyme interactions for diagnostic support.
• To lay the groundwork for scaling up this approach to aid in diagnosing less-understood IMDs.

Main Methods:

• Integrated literature and expert knowledge into machine-readable pathway models, including urine biomarkers and their interactions.
• Applied the framework to visualize clinical data from 16 patients with urea cycle and pyrimidine synthesis disorders on relevant metabolic pathways.
• Utilized expert laboratory scientists to evaluate the diagnostic accuracy of the visualizations.

Main Results:

• The framework successfully visualized patient data, highlighting relevant biomarkers and pathway interactions.
• Diagnostic conclusions derived from the framework matched those from the current diagnostic pipeline for all tested samples.
• Diagnosis was achieved without clinical symptoms or sex information for nine patients, demonstrating the framework's potential.

Conclusions:

• The developed framework effectively integrates metabolic interaction knowledge with clinical data for visualization, aiding in diagnosing complex IMD cases.
• Challenges were identified, requiring resolution before widespread implementation for diagnosing various IMDs.
• Future extensions could incorporate multi-omics data (genomics, transcriptomics) and phenotypic information for enhanced diagnostic capabilities.