In This Issue

Abstract

see Diagnostic yield of array comparative genomic hybridization in adults with autism spectrum disorders Despite its small sample size, a study by Stobbe et al. provides an intriguing glimpse at the extent of discoverable chromosomal abnormalities in a population of autistic adults. Using a whole-genome microarray procedure, the research team performed array comparative genomic hybridization (aCGH) analysis on samples obtained from 24 adult patients seen at the Autism Genetics Clinic, University of Washington, Seattle, from July 2009 to April 2012. aCGH, which can identify small deletions and duplications of a few hundred base pairs, has not previously been used in studies of adults diagnosed with autism spectrum disorders. The technique provided diagnostic information for 5 of 24 patients; 1 additional patient was diagnosed with fragile X syndrome. Variants of unknown significance (VUS) presented in an additional 5 patients. Identification of a specific diagnosis was welcomed by families, the investigators report. Even families who received an inconclusive diagnostic finding expressed gratitude for the opportunity to have an updated genetic evaluation. The hope is that, as variants continue to be explored and catalogued, the prevalence of VUS will diminish. The research team points out that one compelling reason to consider genetic testing of autistic adults is that it may provide information useful in counseling siblings of reproductive age. —Karyn Hede, News Editor see Variations in predicted risks in personal genome testing for common complex diseases In these early days of direct-to-consumer personal genome testing, the reliability of tests varies greatly, as does the viability of companies selling the tests to consumers. In a retrospective analysis, Kalf et al. assessed the predictive ability of three commercial vendors confronted with genomic information for a hypothetical population of 100,000 individuals. The research team calculated disease risk using methodologies provided by the companies and publicly available on their websites or in a published white paper that describes the strategies the companies used for calculating disease risk. The investigators selected six diseases for analysis: age-related macular degeneration, atrial fibrillation, celiac disease, Crohn’s disease, prostate cancer, and type 2 diabetes. This group of diseases, which represent a range of genetic effects, illustrates how single-nucleotide polymorphism selection and differing formulas used for calculating risk affect individual predictions. The findings demonstrated that the testing methodology available in 2008 had limited predictive ability for atrial fibrillation, type 2 diabetes, and prostate cancer and a “considerable (20–27%) probability” of contradictory results from different companies for age-related macular degeneration and Crohn’s disease. The authors note that their study provides what amounts to historical data that could be used in future performance metrics for the industry. In a sign of the volatility of personal genomics as a profit-making enterprise, two of the three companies studied have been acquired by larger life science enterprises, and neither currently offers personal genome testing. —Karyn Hede, News Editor