Jungsul Lee

This paper introduces a deep learning model using multiple instance learning and learning to rank to prioritize variants in rare Mendelian disease genomic testing, achieving high accuracy in identifying disease-causing variants.

The paper develops a system that integrates variant quality control into prioritization to improve the accuracy of identifying disease-causing variants in rare disorders.

The paper evaluates a quality threshold for whole exome sequencing variants to determine when Sanger sequencing confirmation is unnecessary, finding high accuracy with specific quality score and allele fraction thresholds.