Multi-System Genetic Architecture of Hypermobile Ehlers-Danlos Syndrome: Integrating Machine Learning with...

Area of Science:

• Genetics
• Systems Biology
• Machine Learning in Medicine

Background:

• Hypermobile Ehlers-Danlos syndrome (hEDS) lacks a defined genetic basis, hindering diagnosis and treatment.
• Previous research has not fully elucidated the molecular underpinnings of hEDS.
• This study investigates the genetic architecture of hEDS using advanced computational methods.

Purpose of the Study:

• To identify genetic variants associated with hEDS using integrated machine learning and statistical analyses.
• To explore the potential involvement of multiple biological systems in hEDS pathogenesis.
• To decode the genetic architecture of hEDS for improved understanding and potential therapeutic targets.

Main Methods:

• Whole-exome sequencing of 116 subjects from 43 families (86 hEDS patients, 30 controls).
• Analysis of 35,923 rare genetic variants using Random Forest, deep neural networks, and ensemble methods.
• Subject-level Fisher's exact tests with Bonferroni correction for statistical association testing.

Main Results:

• Significant enrichment of variants in collagen biosynthesis, HLA/adaptive immune axis, and mitochondrial respiratory chain pathways in hEDS patients.
• Collagen pathway variants found in 63% of hEDS subjects vs. 17% of controls (p=1.06x10^-5).
• Machine learning models achieved 80% accuracy in classifying hEDS, suggesting complex genetic underpinnings.

Conclusions:

• hEDS genetic etiology may involve variations across multiple biological systems, extending beyond collagen pathways.
• Identified genetic associations require independent validation and functional studies to establish mechanistic relevance.
• Findings provide a hypothesis-generating framework for future hEDS research and potential clinical applications.