1: Structure-Informed Computational Evidence Sharpens BRCA1 Missense Classification

Ramadane-Morchadi L et al., The American Journal of Human Genetics - This episode reviews a study that evaluates how structure-based computational scores (AlphaMissense, FoldX DDG using PDB or AlphaFold2 templates, and RSA) compare with BayesDel for ACMG/AMP PP3/BP4 evidence in classifying BRCA1 missense variants. The authors used MAVE functional data and BRIDGES case-control validation to assess discrimination, evidence strength, and clinical risk association. Findings show AlphaMissense best discriminates functional impact and that combining AlphaMissense with DDG and RSA increases granularity of pathogenicity/benignity evidence. The study highlights that RSA strongly modulates benign evidence and that AlphaFold2 models can serve as DDG templates. Key terms: BRCA1, missense variants, AlphaMissense, DDG, relative solvent accessibility.

Study Highlights:
The authors compared AlphaMissense, FoldX DDG (using PDB and AlphaFold2 models), and BayesDel against MAVE functional scores for 1,638 BRCA1 missense variants and validated major findings in the BRIDGES case-control dataset. AlphaMissense achieved the highest auROC and reduced the fraction of variants in an uninformative score range. DDG predictions add mechanistic granularity and, when combined with AlphaMissense and RSA stratification, permit evidence-strength tiers from supporting to strong. Relative solvent accessibility (RSA) strongly influences benignity evidence (BP4), which is provided mainly for buried or partially buried residues.

Conclusion:
Incorporating structure-informed metrics—AlphaMissense plus DDG and RSA stratification—into ACMG/AMP PP3/BP4 assessment for BRCA1 missense variants improves discrimination, adds evidence-strength granularity, and better aligns computational codes with clinically actionable risk.

QC:
This episode was checked against the original article PDF and publication metadata for the episode release published on 2025-04-15.

QC Scope:
- article metadata and core scientific claims from the narration
- excludes analogies, intro/outro, and music

QC Summary:
- factual score: 10/10
- metadata score: 10/10
- supported core claims: 6
- claims flagged for review: 0
- metadata checks passed: 4
- metadata issues found: 0

Metadata Audited:
- article_doi
- article_title
- article_journal
- license

Factual Items Audited:
- AlphaMissense provides the best discrimination for BRCA1 missense variants, with auROC about 0.93, outperforming BayesDel (0.90) and DDG.
- Combining AlphaMissense with DDG and RSA increases PP3/BP4 evidence granularity (pathogenicity/benignity) beyond single-tool scoring.
- BP4 benign evidence is RSA-dependent and is provided primarily for buried/partially buried residues (RSA ≤60%).
- AlphaFold2 models can serve as effective templates for FoldX DDG predictions in BRCA1 domains, sometimes substituting experimental PDB templates.
- Mutations on surface residues may not receive reliable benign evidence from AlphaMissense or DDG/BayesDel; RSA context is required to interpret BP4 for exposed residues.
- BRIDGES data show structurally unstable BRCA1 variants correspond to higher breast cancer risk (OR > 4).

QC result: Pass.