Detection of copy number variations in epilepsy using exome data

Epilepsies are common neurological disorders and genetic factors contribute to their pathogenesis. Copy number variations (CNVs) are increasingly recognized as an important etiology of many human diseases including epilepsy. Whole exome sequencing (WES) is becoming a standard tool for detecting pathogenic mutations and has recently been applied to detecting CNVs. Here, we analyzed 294 families with epilepsy using WES, and focused on 168 families with no causative single nucleotide variants (SNVs) in known epilepsy-associated genes to further validate CNVs using two different CNV detection tools using WES data. We confirmed 18 pathogenic CNVs, and two deletions and two duplications at chr15q11.2 of clinically unknown significance. Of note, we were able to identify small CNVs less than 10 kb in size, which might be difficult to detect by conventional microarray. We revealed two cases with pathogenic CNVs that one of the two CNV detection tools failed to find, suggesting that using different CNV tools is recommended to increase diagnostic yield. Considering a relatively high discovery rate of CNVs (18 out of 168 families, 10.7%) and successful detection of CNV with <10 kb in size, CNV detection by WES may be able to surrogate, or at least complement, conventional microarray analysis.

WES was performed in a total of 294 families with epilepsy. Then, WES-based CNV detection in 168 families were conducted after excluding 126 families with causative SNVs, and 18 families with pathogenic CNVs were identified. CNVs were detected in two ways: 1) two-step detection: XHMM and subsequent Nord's method (left in a dotted box), and 2) Nord's method targeting 303 epilepsy genes (right in a dotted box). Seventeen and one pathogenic CNVs were detected by methods 1) and 2), respectively. [WES: whole exome sequencing, CNV: copy number variation, SNV: single nucleotide variant, XHMM: eXome Hidden Markov Model]

http://ift.tt/2waSK9z