GWAS meta-analysis: Methodology and application to human meiotic recombination

Human meiotic recombination is critical to successful human reproduction and to maintain genetic diversity. Recombination anomalies are associated with aberrant meiotic outcomes with significant consequences. One important method for studying recombination is genome-wide association studies (GWAS) of recombination phenotypes. Because such studies require nuclear or three-generation family samples that have been genotyped on GWAS chips, the number of suitable datasets is limited. The goal of this dissertation is to develop methods for increasing the available sample sizes for GWAS of recombination phenotypes. We developed two different approaches for increasing sample size. First, we made it possible to include additional family types in the analysis. We developed methods for scoring recombination for half-sibling pedigrees and three generation pedigrees with ungenotyped individuals. Second, we developed a regionally smoothed meta-analysis method for GWAS data, which will allow the combination datasets that have been genotyped on different chips. This method will help increase available sample sizes for recombination studies, but is also applicable to all GWAS studies. The public health significance of this work is that our developments will allow us to find new genes that control recombination and more information about already-known genes. This information can be used for improved treatment and prevention of the consequences of aberrant recombination, including infertility and births with significant chromosomal anomalies.