GWAS Meta-Analysis

This project is still under development. We expect to have a finalized analysis plan around March 2014. Until then this page describes the overarching goal of GSCAN's GWAS project.

The GWAS meta-analysis project is designed to several things.

First, the project will bring together high quality genetic studies of alcohol and nicotine use, resulting in a meta-sample powerful enough to identify very small genetic associations.

Second, the project will take advantage of recent developments in whole genome sequencing and imputation. We will use a large haplotype reference panel to impute to low frequency variants. While the true size of the reference panel remains to be seen, we expect it to contain tens of thousands of individuals, including 1000 Genomes samples.

Third, We will make an imputation server freely available to participating studies (see for a prototype). Imputation and phasing can be a compute and labor intensive process, especially when the haplotype panel is so large (computational burden scales approx. linearly with reference panel size). While the prototype server cannot currently handle large samples or haplotype panels, we will expand the server's computational capability prior to beginning imputation for GSCAN.

Fourth, we will evaluate the same phenotypes as the exome chip and sequencing projects: CPD, DPW, smoking initiation, age of smoking initiation, and pack years.