Abstract
PURPOSE: Gene identification in small families segregating autosomal dominant sensorineural hearing loss presents a significant challenge. To address this challenge, we have developed a machine learning-based software tool, AudioGene v2.0, to prioritize candidate genes for mutation screening based on audioprofiling. METHODS: We analyzed audiometric data from a cohort of American families with high-frequency autosomal dominant sensorineural hearing loss. Those families predicted to have a DFNA2 audioprofile by AudioGene v2.0 were screened for mutations in the KCNQ4 gene. RESULTS: Two novel missense mutations and a stop mutation were detected in three American families predicted to have DFNA2-related deafness for a positive predictive value of 6.3%. The false negative rate was 0%. The missense mutations were located in the channel pore region and the stop mutation was in transmembrane domain S5. The latter is the first DFNA2-causing stop mutation reported in KCNQ4. CONCLUSIONS: Our data suggest that the N-terminal end of the P-loop is crucial in maintaining the integrity of the KCNQ4 channel pore and AudioGene audioprofile analysis can effectively prioritize genes for mutation screening in small families segregating high-frequency autosomal dominant sensorineural hearing loss. AudioGene software will be made freely available to clinicians and researchers once it has been fully validated.