Admixture Mapping Schizophrenia Genes in Oceanic Palau

Investigator: William Byerley, MD
Sponsor: NIH National Institute of Mental Health

Location(s): Palau


Schizophrenia is a devastating psychiatric syndrome that causes untold suffering and losses. Despite differing cultures and environments around the world, the prevalence and presentation of schizophrenia worldwide is similar. Its presentation and course are no different in the Remote Oceanic population of Palau, but the population structure of Palau presents unique opportunities to explore the genetic etiology of this devastating illness. In particular, the population of Palau is of recent origin, founded about 2000 years before present (BP); Palau has undergone at least two substantial bottlenecks, at founding and recently due to disease introduced by European/American contact (approximately 200BP); the population has throughout its history experienced smaller effective population size than most modern societies; and Palau has apparently experienced extensive male-biased gene flow. All of these elements combine to generate linkage disequilibrium (LD) among alleles on autosomal and X chromosomes, which is detectable even at substantial distances on chromosomes. Our principal objective is to identify genes that underlie liability to schizophrenia by using the special features of Palau. In addition to its unique population genetics, Palau has a slightly elevated rate of schizophrenia, 2.77 percent in males and 1.24 percent in females, compared to the 0.5 to 1 percent sex-averaged rate worldwide. There are 156 on-island schizophrenics, of whom 154 have contributed DNA. In addition, we have sampled 495 key relatives. Thus no recruitment is required for this project. Instead, we can focus on linkage and LD analyses to map liability genes. For the former, extended pedigrees and linkage tools are available; only additional genotyping and analyses are required. For the latter, we propose LD analyses on the basis of linkage results and a whole genome scan, which uses a coarse grid of markers. Our preliminary results confirm the likely success of both approaches. Statistical methods will be developed to complement the molecular analyses for the LD component of this study.