Genetic and phenotypic dissection of autism susceptibility

Abstract

Autism is a severe neurodevelopmental disorder characterized by deficits in language and social interaction, and patterns of repetitive and stereotyped behaviors, interests and activities. Evidence indicates that autism has a predominantly genetic etiology, and that as many as fifteen genes may contribute to disease susceptibility. One model suggests autism may result from oligogenic inheritance, with locus heterogeneity, such that different families or individuals possess a different mix of susceptibility alleles. In this dissertation, I present genome-wide linkage studies of autism and traits comprising the aspects of the broader phenotype to identify autism susceptibility loci. I further document detailed molecular and genetic analyses of candidate genes in regions detected by linkage, and in the case of 15q11-q13, as chromosomal duplications found in 1-3% of autism cases. A unifying theme to my dissertation is the focus of genetic studies on genes acting within candidate neurobiological systems suspected of involvement in autism. Genetic analyses include linkage, linkage refinement, construction of detailed linkage disequilibrium (LD) and corresponding haplotype maps across candidate loci, and tests for transmission disequilibrium of single markers and haplotypes. Molecular studies of select candidates aim to identify functional variation on associated alleles; in the absence of association they seek to identify potential rare disease-related variants considering for example evolutionarily conserved sequence. I hypothesize that there are allelic variants, which underlie genetic linkage and/or association to autism and related traits, and these contribute to autism susceptibility through both direct and interactive effects. The goal of this study is to dissect the genetic etiology of autism by leveraging trait-based phenotypic subsets of autism using the approaches and tools I have outlined here.