**3. Results**

**2.4. Next-generation sequencing**

392 Next Generation Sequencing - Advances, Applications and Challenges

**2.5. Data analysis**

TheTruseqcustomampliconlibrarypreparationkitandtheMiSeqIlluminasequencerplatform (Illumina Inc.) wereusedto create the sequencing library andperformresequencing respective‐ ly. All steps were performed in-house according to the manufacturer's instructions [27, 28]. Library preparation was performed by enrichment of the target regions using an ampliconbased multiplex polymerase chain reaction (PCR) method. Here, a custom amplicon tube (CAT) containing upstream and downstream oligonucleotides specific for the target regions was hybridized to the unfragmented gDNA samples in a 96-well plate. Unbound oligonu‐ cleotides were then removed by a series of wash steps using manufacturer supplied reagents. A proprietary extension–ligation mix containing DNA polymerase and ligase (Illumina Inc.) extended and ligated the upstream bound oligonucleotide through the targeted region to the 5′ end of the downstream oligonucleotide. The resulting extension–ligation products contain‐ ing the targeted genomic region flanked by common sequences required for amplification were then amplified by standard PCR on a thermal cycler. The amplicon size (250 bps), the number of amplicons in the CAT (632 amplicons) and the type of input DNA (high quality) determined the number of PCR cycles (*n* = 24). The PCR reaction incorporated two unique, sample-specific, multiplexing index sequences (barcoding) that would later be used by the alignment software (MiSeq reporter) to identify individual samples following library pooling, and common adapters required for cluster generation. PCR products were purified by AMPure XP beads (Beckman Coulter, Lane Cove, NSW, Australia) and the quantity of each library was normal‐ ized by an integrated bead-based method. Equal volumes of the normalized libraries were

then combined, diluted in hybridization buffer (Illumina Inc.) and heat denatured.

for bases with insufficient coverage and validate variants of clinical significance.

The University of California, Santa Cruz (UCSC), genome browser (http://genome.ucsc.edu) was used for variant analysis and variants were cross-checked against databases including the NHLBI-Extended Sequencing Project (ESP), 1000 Genomes Project Database [31] and the Database of Single-Nucleotide Polymorphisms (dbSNP, http://www.ncbi.nlm.nih.gov/SNP/). Bioinformatic tools, Sorting Intolerant From Tolerant (SIFT, http://sift.jcvi.org/) [32], Polymor‐ phism Phenotyping-2 (PolyPhen-2, http://genetics.bwh.harvard.edu/pph2/) [33] and Mutation

The MiSeq Illumina instrument was used to resequence the pooled library by paired-end sequencing. The DNA library was immobilized to the single-use glass-based MiSeq flow cell through the adapter sequences. Bridge PCR amplification then generated clusters of clonal copies of each DNA molecule. These templates were then sequenced using platform-specific reversible dye terminator sequencing-by-synthesis chemistry. Sequence alignment to the reference genome (GRCh37/hg19) was performed using on-instrument software (MiSeq reporter software, Illumina Inc.) that aligned the reads in BAM format and outputted variant calls in.vcf files. Variant calls were generated using ANNOVAR software (http://www.open‐ bioinformatics.org/annovar) [29] with an acceptance threshold Q-score of 30, corresponding to a 1:1000 error rate and genomic datasets were viewed using the Integrative Genomics viewer (IGV) (www.broadinstitute.org/igv/) [30]. Sanger sequencing was performed to provide data
