Genomewide association study of cocaine dependence in two populations Abstract: Cocaine dependence (CD) is highly costly to society;genetic epidemiologic studies support moderate heritability for this trait. We have completed a genomewide linkage study for this trait, and several candidate gene associations have been proposed. CD cuts across society;however, minority populations, and specifically African-Americans (AAs), though affected by CD at high rates, are particularly understudied. We have been very successful at collecting substance-dependent (SD) EA and AA unrelated subjects and families, and based on that sample, we propose a genomewide association study (GWAS) of CD in both AA and EA samples. Affected subjects were assessed with the rigorous and reliable SSADDA instrument. Because there is substantial comorbidity of nicotine dependence (ND) (which is also moderately heritable) with CD, we propose genomewide association analysis of ND as a secondary aim (as we have similarly, and successfully, done in prior linkage studies). Our initial sample size of a total of 3500 subjects with CD (2000 AA and 1500 EA), and 3075 subjects with ND (1575 AA and 1500 EA) (with many individuals comorbid for the two traits), will provide adequate power to detect novel risk loci for both traits, when compared to control samples composed of subjects recruited by ourselves, and our collaborator Ming D. Li, with the AA control sample augmented with genotypes already in the public domain (PD). We have assembled >1400 AA controls to genotype (979 AA subjects already been genotyped on the Affymetrix 6.0 array as part of project GAIN with data publicly available);and >1600 EA controls. We propose to genotype AA subjects on the Affymetrix 6.0 array, and EA subjects on the Illumina 1M array. This will provide best available coverage in each population, and at the same time allow best possible comparability with existing PD samples. For replication, we have assembled a collection of an additional >2350 CD subjects (including 1100 SSADDA-assessed subjects not required for the initial GWAS) and >1400 controls, to be studied with a custom 1536-SNP array. Additionally, Dr. Li has agreed to evaluate independently any ND findings in his own sample. Finally, we will conduct a high-resolution CNV genomewide assessment using NimbleGen HD2 high- density oligonucleotide microarrays on a subset of 500 cases and 500 controls. This will allow us to consider an additional source of polymorphic variation at much higher resolution, and with greater uniformity across the genome, than can be obtained even with modern high-density genotyping arrays. This will be a pioneering CD GWAS, and is particularly noteworthy for its focus on two major substance dependence traits in a minority population. We therefore expect to obtain invaluable new insights into CD in the major US populations. Follow-up plans include replication in an independent sample and high-throughput sequencing of implicated genomic regions (the latter separate from the present proposal).
Narrative The major goal of this project is to identify risk genes and variants associated with cocaine dependence, using the genomewide association study technique. A secondary goal is to identify risk genes and variants for nicotine dependence. Finding these risk genes will aid in understanding, prevention, and eventually, treatment, of these disorders.
Showing the most recent 10 out of 52 publications
