Genome-wide association studies (GWAS) of the electrographic QT-interval, an intermediate trait that impacts the risks of long QT syndrome and sudden cardiac death, have identified 68 independent variants at 35 loci, explaining 8% of the phenotypic (20% of the additive) variance. Nevertheless, the identity, function and mechanisms of action of the underlying DNA sequence variants and genes remain unknown, and are major impediments for understanding the molecular structure and functional architecture of this complex phenotype. We hypothesize that the majority of functional trait variants are polymorphic, non-coding and perturb transcription of a specific gene by altering the functions of their cis-regulatory elements. We propose a research paradigm for systematically identifying these non-coding trait variants, the regulatory functions they disrupt and the specific genes whose functions are altered at each quantitative trait locus. We will utilize integrative statisticl genetic, computational, molecular genetics and cellular approaches for elucidating the underlying mechanisms, using QT interval as a model 'system'.
Our specific aims are: (1) to perform high- resolution mapping of GWAS signals to identify all polymorphic (>1%) and rare variants at loci that modulate the QT-interval; (2) to conduct in silico and in vitro analysis to predict and prioritize all cardiac regulatory (enhancer, silencer, insulator) elements and their cognate DNA-binding proteins; and, (3) to identify trait variants, genes and their mechanisms of genetic action. The overall goals are to improve the molecular genetic and mechanistic understanding of multifactorial traits for applications to other complex phenotypes.
The electrocardiographic QT-interval, an intermediate trait for arrhythmias and sudden cardiac death, has 68 predominantly non-coding variants at 35 loci explaining 8% of the phenotypic (20% of the additive) variance upon meta-analyses in >76,000 individuals of European and >11,000 individuals of African ancestry: nevertheless, the molecular details remain unknown. We propose a program of research for systematically identifying the functional non-coding quantitative trait variants, the regulatory function they disrupt and the specific gene whose functions are altered, at each locus. We utilize integrative statistical genetic, computational, molecular genetics and cellular approaches for understanding the underlying genetic mechanisms in this model trait with the overall goal to improve the molecular understanding of all complex phenotypes.