Complementary Roles of Structure and Variant Effect Predictors in RyR1 Clinical Interpretation.

Area of Science:

• Genetics and Molecular Biology
• Biochemistry
• Computational Biology

Background:

• RyR1-related disorders stem from RYR1 gene variants, presenting diverse phenotypes.
• Interpreting RYR1 variants is challenging due to gene length and variant mechanisms.
• Current variant effect predictors (VEPs) have limited performance and inherent biases.

Purpose of the Study:

• To evaluate the efficacy of 70 VEPs for RYR1 missense variant classification.
• To introduce a novel protein structure-based metric, Spatial Proximity to Disease Variants (SPDV).
• To aid in the clinical interpretation of RYR1 variants of uncertain significance.

Main Methods:

• Evaluated 70 VEPs using pathogenic and benign RYR1 missense variants.
• Introduced SPDV, a metric based on 3D clustering of pathogenic mutations.
• Determined ACMG/AMP PP3/BP4 classification thresholds for SPDV and top VEPs.

Main Results:

• Existing VEPs showed variable performance; those trained on clinical data had inflated performance due to circularity.
• VEPs minimizing training bias showed limited performance, potentially missing gain-of-function variants.
• SPDV demonstrated utility in assigning PP3/BP4 evidence levels to uncertain RYR1 variants.

Conclusions:

• A novel protein structure-based approach (SPDV) offers an orthogonal strategy to existing VEPs.
• SPDV aids in the diagnostic process for RyR1-related diseases.
• This method helps overcome limitations of current computational tools for variant interpretation.