Association testing to detect gene-gene interactions on sex chromosomes in trio data

• Main Authors: Yeonok eLee, Debashis eGhosh, Yu eZhang
• Format: Article
• Language: English
• Published: Frontiers Media S.A. 2013-11-01
• Series: Frontiers in Genetics
• Subjects: Sex Chromosomes; gene-gene interaction; binary traits; generalized linear mixed effect model; logistic model; trio data
• Online Access: http://journal.frontiersin.org/Journal/10.3389/fgene.2013.00239/full

Autism Spectrum Disorder (ASD) occurs more often among males than females in a 4:1 ratio. Among theories used to explain the causes of ASD, the X chromosome and the Y chromosome theories attribute ASD to X-linked mutation and the male-limited gene expressions on the Y chromosome, respectively. Despite the rationale of the theory, studies have failed to attribute the sex-biased ratio to the significant linkage or association on the regions of interest on X chromosome. We further study the gender biased ratio by examining the possible interaction effects between two genes in the sex chromosomes. We propose a logistic regression model with mixed effects to detect gene-gene interactions on sex chromosomes. We investigated the power and type I error rates of the approach for a range of minor allele frequencies and varying linkage disequilibrium between markers and QTLs. We also evaluated the robustness of the model to population stratification. We applied the model to a trio-family data set with an ASD affected male child to study gene-gene interactions on sex chromosomes.