The landscape and clinical impact of cryptic structural variation in autism and related neuropsychiatric disorders. H. Brand1,2,3,8, V. Pillalamarri1,8, R. Collins1, S. Eggert1,4, M. Stone1, I. Blumenthal1, C. O'Doushlaine3, E. Braaten2, J. Rosenfeld5, S. Mccarroll3,4,6, J. Smoller1,2,7, A. Doyle1,2,3,6,7, M. Talkowski1,2,3,7 1) Center for Human Genetic Research, Massachusetts General Hospital, Boston, MA; 2) Departments of Psychiatry, Neurology, Harvard Medical School; 3) Program in Medical and Population Genetics, Broad Institute of MIT and Harvard; 4) Department of Genetics, Harvard Medical School; 5) Signature Genomic Laboratories, Perkin Elmer Inc., Spokane, WA; 6) Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard; 7) Psychiatric and Neurodevelopmental Genetics Unit, Massachusetts General Hospital; 8) Co-First Author.

   Rare, loss-of-function (LoF) structural variation (SV) represents a major component of the genetic risk in autism spectrum disorder (ASD) and other neuropsychiatric disorders (NPD), however much of the etiology remains unexplained. Most SV studies have focused on copy number variants (CNV) using microarray technology, which is blind to the presence of balanced chromosomal abnormalities (BCAs). Indeed, the diagnostic yield of BCAs in ASD is unknown as CNV analysis is the currently recommended first-tier screen, and there is no standardized clinical approach in other NPDs such as early onset schizophrenia or bipolar disorder. We have previously shown that cytogenetically visible BCAs represent a unique class of highly penetrant LoF variation, however the impact of cryptic BCAs is uncharacterized as they remain intractable to all conventional technology other than deep whole-genome sequencing (WGS). Here, we delineate the full mutational spectrum of SV in the initial 250 families in an SV sequencing study of ASD or other common NPDs (ADHD, bipolar disorder, major depressive disorder) using large-insert jumping library WGS with a median insert of 3.7 kb a genome-wide average insert coverage of 64X. We discovered a spectrum of SVs ranging from karyotypically visible and clinically significant BCAs to highly complex cryptic chromothripsis. We also find numerous duplications as delineated by microarray that are duplicated insertions or complex rearrangements that can also cause LoF at the insertion site(s). On average, we find 61 cryptic CNVs and 48 cryptic BCAs per genome at this resolution. We also find that, on average, cryptic SVs detectable in our study rearrange 3.6 Mb of the genomes, disrupting 98 protein coding genes, and result in at least 19 additional LoF variants per individual. Using a convergent genomic approach of assessing CNV and exome burden from over 50,000 additional subjects, we estimate the clinically significant yield of cryptic SVs to be >5%, a potentially remarkable component of the disease variance that remains uncharacterized. Clinically significant SVs included de novo inversion of a known driver of a microdeletion syndrome (2q23.1), complex chromothripsis, and discovery of novel ASD genes such as UBE2F. These data suggest that the clinical yield from cryptic SVs in ASD and NPDs is substantial, and that SV detection warrants consideration in diagnostic testing for early onset NPDs. Analyses of larger cohorts is ongoing.

You may contact the first author (during and after the meeting) at