Using whole exome sequencing to walk from clinical practice to research and back again.

W hole exome sequencing (WES) is currently used to identify the genetic causes of many diseases, especially monogenic disorders. Ng et al, 1 in 2009, completed the first proof-of-principle study demonstrating the feasibility of using exome sequencing to identify causal variants for diseases, specifically Freeman-Sheldon syndrome. Within 2 years, there was a marked increase in publications presenting WES data, and the pace continues to accelerate (Figure). In 2010, WES began to be used for clinical diagnoses, particularly for mendelian disorders. In early 2011, Worthey et al 2 used exome sequencing to facilitate clinical diagnosis and modify treatment in a single case. Despite many of the successes resulting from exome sequencing, more than half of the approximately 7000 known or suspected mendelian disorders have not yet been discovered, 3 which highlights the need for more genetic, mechanistic, and clinical studies, particularly if the data are to be used clinically. Moreover, as our knowledge of the genome increases, examples of some of the complexities associated with genotypic-phenotypic relationships further substantiate the need for both additional genomic annotation and many more sequenced genomes with phenotypic information. Some of these complexities include the following: (1) Variants in the same genes may lead to different clinical manifestations or phenotypes; (2) what appear to be similar phenotypic observations may result from different causal disease variants operating through distinct pathophysiological mechanisms; and (3) the recent ENCODE (Encyclopedia of DNA Elements) papers, which suggest that up to 80% of the human genome is functional. In this issue of Circulation, Crotti and colleagues 5 use a clever strategy to investigate the cause of cardiac arrhythmias, ventricular fibrillation, and cardiac arrest in 2 unrelated pro-bands who were negative for variants in KCNQ1, KCNH2, SCN5A, KCNE1, and KCNE2, which are the genes most commonly involved in long-QT syndrome (LQTS). Specifically, they began by performing WES on the 2 unrelated probands and their healthy parents. As with other reported studies, they looked only at rare variants. Analysis of the WES results for the 2 probands uncovered de novo mutations in either CALM1 or CALM2, which both encode for calmodulin. At this point, the team faced the problem all investigators and clinical teams face when a new and potentially relevant variant(s) is found: Is it causal? Here, the team used a validation strategy that other groups have also used. 6 They conducted follow-up genetic screening of the calmodulin genes (CALM1, CALM2, and CALM3) …