Phased allele-specific expression analysis in integrated whole-exome and mRNA sequencing study in a family with non-random X chromosome inactivation. S. Szelinger1,2,3, V. Narayanan2,4, J. J. Corneveaux1,2, I. Schrauwen1,2,5, A. L. Siniard1,2, A. A. Kurdoglu1,2, I. Malenica1,2, K. M. Ramsey1,2, M. J. Huentelman1,2, D. W. Craig1,2 1) Neurogenomics Division, Translational Genomics Research Institute, Phoenix, AZ; 2) Center for Rare Childhood Disorders, Translational Genomics Research Institute, Phoenix, AZ; 3) Molecular and Cellular Biology Interdisciplinary Graduate Program, School of Life Sciences, Arizona State University, Tempe, AZ; 4) Pediatric Neurogenetics Center, Barrow Neurological Institute, Phoenix, AZ; 5) Department of Medical Genetics, University of Antwerp, Antwerp, Belgium.

   We describe a diagnostic framework for high-throughput sequencing data to determine magnitude and mode of non-random X chromosome inactivation (XCI) in a clinical female patient diagnosed with mild cognitive impairment, using integrated, family trio-based whole-exome and mRNA sequencing. Female carriers of deleterious genetic variations causing X-linked disease can present with heterogeneous phenotypes due to XCI. Random XCI in females results in an approximate equal ratio of cells expressing maternal or paternal X chromosome genes. When XCI is non-random, this ratio can vary from mildly skewed XCI (e.g. 70:30) to completely skewed XCI (e.g. 0:100). Current screening methods in clinical practice to detect biased X chromosome inactivation rely on indirect observation of expression by DNA methylation status of a single genetic locus. Truseq exome sequencing of the trio on two Hiseq2000 lanes resulted in a mean base coverage of the captured exonic regions of 86.25X. TruSeq mRNA sequencing of the trio on a single lane of a HiSeq2000 run resulted in an average of 116 million paired, mapped reads. Heterozygous SNP genotypes on chromosome X were phased in the patient by following the transmission of alleles identified in the parental exome SNP set. Phasing uncovered the paternal X as the source of a de novo, interstitial, 1.7 Mb deletion on the Xp22.31 locus that is recurrently been identified as the source of mental retardation phenotype in X-linked diseases. Allele-specific mRNA expression analysis of phased SNP alleles identified 80:20 non-random XCI that favored the expression of the maternal, cytogenetically normal X and compared well to the 85:15 ratio determined by the standard DNA methylation assay. In conclusion, the integrated whole-exome and mRNA-seq data suggests that the deleterious effect of the deletion on the paternal copy may be offset by skewed XCI that favors expression of maternal X resulting in the mild phenotype. These results demonstrate the utility of the combined approach of high-throughput DNA and RNA sequencing, phase-by-transmission allelic expression analysis in understanding the molecular basis of X-linked disorders with moderate phenotypes.

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