Andrew G. Sharo

GWAC is a machine learning method that identifies functional variants in data-constrained species using sequence-based features, performing nearly as well as CADD in humans.

The paper uses museum genomics to compare historical and contemporary steelhead populations, finding minimal introgression and a decline in migration-associated haplotypes over time.

The paper assesses improvements in genomic variant classification accuracy over time in ClinVar and HGMD by analyzing misclassification rates in inborn errors of metabolism across ancestry groups.

StrVCTVRE predicts the pathogenicity of human genome structural variants using a random forest classifier, achieving 90% sensitivity while filtering half of SVs.

This study evaluates the Full Genome Analysis method for diagnosing rare monogenic disorders, finding a 40% diagnostic yield in clinical cases.