Suzanne de Bruijn

235 Exploring the missing heritability in subjects with hearing loss and EVA Medical history was taken from all participants and special attention was paid to non- genetic causes of HL. Results of pure tone, speech, and brainstem evoked response audiometry, performed in a sound-attenuated booth, were collected. Air and bone conduction pure tone thresholds were determined for frequencies ranging from 0.25 to 8 kHz. Threshold estimates based on brainstem evoked response audiometry were used when pure tone audiometry was not available. Individuals were considered affected when pure tone thresholds for at least three frequencies were above the frequency- specific 95 th percentile of age- and sex-specific thresholds (ISO 7029:2017) for the best hearing ear. In the Netherlands, routine newborn hearing screening is carried out by the detection of transient evoked otoacoustic emissions. 24 When available, these data were used to determine whether the HL was congenital. Previously performedCT andMRI scanswere retrieved and reassessedby an experienced neuroradiologist (S.A.H.P.). An EVA was defined as a vestibular aqueduct that measured ≥2mmat the operculumand/or ≥1mmat themidpoint 25 , in accordance with previously published reports on this topic. 22,23 Analyses of pair-wise differences between patient groups were performed with R (R Foundation) using multivariate linear regression analysis (using lsmeans 2.3.0) with a correction for multiple comparisons using the Holm method. 26 Next generation sequencing and variant interpretation Genomic DNA was isolated from peripheral blood lymphocytes and analyzed by molecular inversion probe (MIP) sequencing, whole exome sequencing (WES) or whole genome sequencing (WGS) ( Table S2 ). For WES, exome enrichment was performed using the Agilent SureSelect Human All Exome V4 or V5 kits according to the manufacturer’s instructions. Subsequently, sequencing was executed on an Illumina HiSeq system by BGI Europe (Copenhagen, Denmark), with a minimal coverage of 20x for 93.77% of the targets and an average coverage of >100 reads. Read mapping along the hg19 reference genome (GRCh37/hg19) and variant calling was performed using BWA V.0.78 27 and GATK HaplotypeCaller V.3.3 28 , respectively. An in-house developed pipeline was used for variant annotation and copy number variant (CNV) detection was performed using CoNIFER V.0.2.2.3 29 . WGS was performed by BGI (Hongkong, China) on a BGISeq500 using a 2x 100 bp paired end module, with a minimal median coverage of 30-fold per genome. Read mapping (GRCh37/hg19) and variant calling was performed as described for WES. Structural variants (SVs) were called using the Manta Structural Variant Caller V.1.1.0 (SV detection based on paired end and split read evidence) 30 and CNVs using Control-FREEC (CNV detection based on alterations in read depth. 31 MIP design, sequencing and data analysis were performed as previously described. 32,33