Targeted Exome Sequencing for Hereditary Hearing Loss

DNA samples obtained from the proband (II-1), his younger brother (II-2), his father (I-1) and his mother (I-2) were sequenced using target exome-based next-generation sequencing. Roche NimbleGen’s (Madison, United States) custom Sequence Capture Human Array was used to designed to capture targeted sequence, covering all exons and flanking sequence (including the 100 bp of introns) of 127 genes which is associated hereditary hearing impairment (Supplementary Table 1). The details of targeted next generation sequencing have been described in Supplementary Table 2. The 127-gene panel achieved a total of 619.167 kb of targeted sequence, covering 2,268 exons and flanking sequence. An average of 2022981 reads per sample was acquired, with approximately 85% mapping to their targets. The average mean depth for the targeted regions was 311.3 ± 56.7; 97.5 ± 0.1% of the covered exons had ≥ 30 reads. Average depth and coverage of Target genes has been described in Figure 2A. Figure 2B, is showing read depth at this causal variant in BAM file across ILDR1 locus. The procedure for preparation of libraries was consistent with standard operating protocols published previously (Yang et al., 2018 (link)). According to the standard protocol, simultaneously we sequenced 30 samples on Illumina HiSeq 2500 Analyzers (Illumina, San Diego, United States) for each pooling batch for 90 cycles (specially designed rare disease screening). We applied Illumina Pipeline software (version 1.3.4) to generate the raw data which is later used for Bioinformatic analysis. We extract the clean reads from the raw reads by using already established filtering criteria (Wei et al., 2014 (link)). Then, we selectively using at least 90 bp long clean reads for aligning to the human reference genome (Build 37) of NCBI database by using Burrows Wheeler Aligner (BWA). BWA, a multi-vision software package, generating the output file in bam format. After that, target region coverage, sequencing depth, SNP/InDel/CNV detection has been analyzed by using the bam data. Next, SOAPsnp software, Sam tools pileup software and Bioinformatic computational framework were established for identifying SNP, InDels and CNVs. Filtering criteria for a SNP or InDel has been set with at least 10 reads and >20% of total reads. SNPs are filtered out and selected for further interpretation if the frequency of the SNPs is <0.05 in dbSNP, HapMap, 1000 Genomes database, the 100 healthy reference samples (same ethnic origin with similar age and sex range) sequenced in this study. The comprehensive and detailed method of variant interpretation has been described in Figure 3A.