Domain-Specific Computational, Functional and Structural Methods Enable Interpretation of BRCA1 BRCT Variants of...

Area of Science:

• Genomics and Bioinformatics
• Cancer Genetics
• Molecular Biology

Background:

• Germline BRCA1 and BRCA2 variants are primary drivers of hereditary breast and ovarian cancers.
• Thousands of variants of uncertain significance (VUS) complicate genetic test interpretation and clinical decisions.
• The majority of BRCA VUS are missense variants, necessitating accurate pathogenicity assessment.

Purpose of the Study:

• To enhance the accuracy of BRCA1 VUS pathogenicity evaluation.
• To utilize computational, functional, and structural methods for variant classification.
• To address the clinical challenge posed by numerous BRCA1 missense VUS.

Main Methods:

• Structural analysis identified RING and BRCT domains as variant hotspots.
• A computational classifier was developed focusing on the BRCT domain.
• Machine learning models integrated nine in silico tools for pathogenicity prediction.
• Functional assays assessed phosphopeptide binding and protein levels for select VUS.

Main Results:

• The BRCT domain was identified as a key region for missense variants and VUS.
• A computational classifier integrating nine in silico tools was developed for BRCT VUS.
• Twenty-two BRCA1 VUS were functionally assessed, showing varied phosphopeptide binding and protein levels.
• Computational structural modeling provided insights into VUS interactions and structural impacts.

Conclusions:

• In silico and functional data support the classification of BRCA1 BRCT VUS.
• Domain-specific computational approaches are effective for characterizing missense variants in multi-domain genes.
• This study enhances the accurate classification of BRCA1 VUS, aiding clinical decision-making.