Cracking the Code: Genotype-Phenotype Correlation Models in Sarcoglycanopathies.

Area of Science:

• Genetics
• Molecular Biology
• Neurology

Background:

• Sarcoglycanopathies are severe limb-girdle muscular dystrophies (LGMD) often caused by missense variants.
• Predicting variant effects on protein function is challenging, hindering genotype-phenotype correlations.
• Accurate genotype-phenotype correlations are crucial for understanding disease progression and management.

Purpose of the Study:

• To establish precise genotype-phenotype correlations for limb-girdle muscular dystrophy types R3, R4, and R5 (LGMDR3, LGMDR4, LGMDR5).
• To develop predictive models for variant impact on sarcoglycan maturation and function.
• To improve diagnostic accuracy and prognostic capabilities for patients with sarcoglycanopathies.

Main Methods:

• Analysis of the largest sarcoglycanopathy cohort (n=541) with detailed clinical data.
• Development of predictive models incorporating variant impact on protein secondary structure, functional domains, conservation, and interactions.
• Classification of variants based on predicted disruption of membrane trafficking to assess phenotype severity.

Main Results:

• A model for LGMDR3 achieved 89% predictive power and 86.7% balanced accuracy by analyzing variants in specific domains.
• Models for LGMDR4 and LGMDR5 demonstrated high predictive power (80% and 90% respectively) and low false-negative rates.
• An open-access predictive tool was developed for clinical and research use.

Conclusions:

• The study establishes robust genotype-phenotype correlations for sarcoglycanopathies.
• Findings will improve patient prognosis and management strategies.
• Enhanced correlations will facilitate more effective clinical trial recruitment and design.