Addiction is a common and costly public health problem, and tobacco use is the leading cause of preventable deaths in the US, leading to an estimated 438,000 premature deaths per year. Nicotine is the substance in tobacco which is responsible for its addictive properties. The goal of our project is to identify gene variants that contribute to nicotine dependence, using both linkage and association studies of subjects from the Icelandic population. To study the genetics of nicotine dependence we have recruited over 5,000 smokers and re-phenotyped them using questionnaires and by interviews. To isolate susceptibility variants we have fine mapped regions linked to nicotine dependence through fully multipoint allele sharing linkage analysis that does not assume a particular inheritance model. Significant linkage peaks with high information content are ultrafine-mapped with hundreds of markers to define the underlying LD structure. The regions have been assessed in a case-control analysis to look for significant haplotype association to nicotine dependence. Genes found in Iceland using this approach are tested for their impact in outside populations already collected by scientific collaborators. The extensive phenotype data on the nicotine dependent patients and relatives will be used in more careful genotype-phenotype correlations as a way to understand their role in nicotine dependence and its co-morbidities and treatment for this common and intractable problem, in humans representing most important risk factors for numerous diseases. The project is also conducting genome-wide association studies of smoking behavior in a large sample of smokers. The studies have already lead to the identification of three sequence variants correlating with ND. All variants also confer risk of lung cancer, underscoring the public health importance of understanding the genetics of nicotine dependence. In the continuation we plan to increase sample size for GWA studies in Iceland using long-range phasing approaches and joint analyses of large datasets. We will conduct pool sequencing for the discovery of less frequent variants that will subsequently be typed on a set of 2,500 samples that have been chosen to optimize their usefulness for propagating human sequence information into the rest of the Icelandic population using long-range-phasing approaches. The pools will be enriched in samples from carriers of variants detected in GWA studies, and high-risk haplotypes detected by long-range phasing analysis with a focus on linkage regions. In addition we will pay special attention to copy number variants in our approaches. To validate variants in identified in this manner in other populations we will rely on large sets of foreign ND and smoking behavior case-control samples, and a larger set of samples rich in information on both smoking-related diseases and psychiatric disorders.