A Contemporary Review of AudioGene audioprofiling: A machine-based candidate gene prediction tool for autosomal dominant nonsyndromic hearing loss
✍ Scribed by Michael S. Hildebrand; Adam P. DeLuca; Kyle R. Taylor; David P. Hoskinson; In Ae Hur; Dylan Tack; Sarah J. McMordie; Patrick L. M. Huygen; Thomas L. Casavant; Richard J. H. Smith
- Publisher
- John Wiley and Sons
- Year
- 2009
- Tongue
- English
- Weight
- 347 KB
- Volume
- 119
- Category
- Article
- ISSN
- 0023-852X
No coin nor oath required. For personal study only.
✦ Synopsis
Significant hearing loss (25 dB) is prevalent in the adult population (15%-20%) and affects approximately 50% of individuals 80 years of age or older. 1 It is estimated that 15% of inherited hearing impairment is accounted for by autosomal dominant nonsyndromic forms of hearing loss (ADNSHL). To date, 57 loci for autosomal dominant deafness have been mapped to chromosomal regions, and 23 causally related genes have been identified at these loci (http://webh01.ua.ac.be/hhh/; May 8, 2009). 2 In most cases the hearing loss is sensorineural and nonsyndromic.
Phenotype-genotype relationships in ADNSHL have been extensively studied. 3 Based on these data, it is clear that some correlations are very robust, such as the low-frequency audioprofile associated with WFS1-related hearing loss (DFNA6/14/38), 4 whereas other correlations are more difficult to define. Autosomal dominant highfrequency hearing loss, for example, can be caused by mutations in a large number of different genes including KCNQ4 (DFNA2), DFNA5 (DFNA5), COCH (DFNA9), and POU4F3 (DFNA15). By analyzing additional audiometric data at each of these loci, however, it may be possible to cluster genes that cause high-frequency hearing loss into definable subclusters. To determine whether these genes fall into phenotypically defined subclusters, multiple regression studies of threshold data can be conducted taking into account age and/or select frequencies. This type of audioprofile subclustering could potentially increase the efficiency of gene identification in small families segregating ADNSHL.
To address this challenge we have developed and reported on a computer program, AudioGene, that uses a machine-learning approach to analyze audioprofiles as a method of prioritizing genes for mutation screening in small families segregating ADNSHL (Fig. 1). 5 The purpose of this communication is to describe improvements to our software and to introduce a new website (http://audiogene.eng. uiowa.edu/audiogram) that we have developed to allow physicians access to the AudioGene version 3.0 software.