Matthew Hayden

The paper compares the functionality and pathogenicity of 5' splice site GC>GT and GT>GC variants using a meta-analysis and functional assay, finding that GC>GT variants are largely non-pathogenic.

The paper evaluates SpliceAI's performance in predicting the functional impact of GT>GC 5' splice site variants using experimental data and finds it is not fully accurate.

The paper estimates that 15-18% of canonical GT 5' splice sites can generate wild-type transcripts when mutated to GC, with implications for disease and splicing mechanisms.