MitoPhen database: a human phenotype ontology-based approach to identify mitochondrial DNA diseases

Area of Science:

• Genetics and Genomics
• Rare Diseases
• Bioinformatics

Background:

• Mitochondrial disorders are challenging to diagnose due to overlapping clinical presentations with other genetic conditions.
• Despite advances in genetic testing, phenotypic and genetic heterogeneity complicate timely and confident molecular diagnoses.
• A systematic approach to describing variant-specific phenotypic spectra is needed to improve diagnostic accuracy.

Purpose of the Study:

• To establish the MitoPhen database, a comprehensive resource for cataloging clinical phenotypes associated with mitochondrial DNA (mtDNA) mutations.
• To enable hypothesis-free definition of mtDNA clinical syndromes and explore heteroplasmy-phenotype relationships.
• To provide a tool for objective clinical comparison of new patients with existing data using human phenotype ontology (HPO) terms.

Main Methods:

• Curated clinical phenotypes from 6688 individuals with 89 pathogenic mtDNA mutations.
• Collected 26,348 human phenotype ontology (HPO) terms to build the MitoPhen database.
• Developed HPO-based phenotype similarity scores to distinguish between mitochondrial and non-mitochondrial rare disease patients.

Main Results:

• The MitoPhen database facilitates hypothesis-free definition of mtDNA clinical syndromes and identification of under-recognized phenotypes.
• HPO-based phenotype similarity scores successfully distinguished between patients with mtDNA diagnoses and those with nuclear genetic diagnoses (FDR <10%, sensitivity 80%).
• The database serves as a publicly available reference for objective clinical comparisons.

Conclusions:

• The MitoPhen database and HPO-based similarity scores offer a systematic method to improve the diagnostic confidence and timeliness of mitochondrial disorders.
• This approach aids in understanding heteroplasmy-phenotype correlations and defining novel clinical syndromes.
• Further enrichment of the database with more patient data will enhance diagnostic predictions for rare variants.