Michael T. Parsons

The paper develops machine learning models to predict pathogenicity of BRCA1 and BRCA2 variants using cancer genomic profiles, improving classification of variants of uncertain significance.

The paper uses case-control data from over 400,000 women to classify BRCA1 and BRCA2 variants, finding high consistency with ClinVar annotations for 785 previously unclassified variants.

The paper refines ACMG/AMP variant classification criteria for BRCA1 and BRCA2 by calibrating evidence strength and improving classification consistency through expert panel review.

The paper introduces SAI-10k-calc, a tool that extends SpliceAI to predict splicing aberrations and their effects, achieving high sensitivity and specificity for variant impact prediction.

The paper refines ACMG/AMP splicing variant classification codes using empirical evidence to improve consistency in variant pathogenicity classification.

The paper assesses breast cancer risks linked to rare missense variants in five susceptibility genes using logistic and mixture models, finding varying risk contributions across genes.

The paper demonstrates federated analysis of BRCA1 and BRCA2 variants in a Japanese cohort to interpret variants of uncertain significance without transferring sensitive data.