PedGenie: meta genetic association testing in mixed family and case-control designs

Meta statistics (chi-squared tests, odds ratios, and confidence intervals) were calculated using formal Cochran-Mantel-Haenszel techniques. Simulated data from unrelated individuals and individuals in families were used to illustrate meta tests and their empirically-derived p-values and confidence intervals are accurate, precise, and for independent designs match those provided by standard statistical software.PedGenie yields accurate Monte Carlo p-values for meta analysis of data across multiple studies, based on validation testing using pedigree, nuclear family, and case-control data simulated under both the null and alternative hypotheses of a genotype-phenotype association.PedGenie allows valid combined analysis of data from mixtures of pedigree-based and case-control resources. Added meta capabilities provide new avenues for association analysis, including pedigree resources from large consortia and multi-center studies.In the study of common diseases and genes with modest effects, large consortium and multi-center efforts hold the promise of increased power to detect associations, but present analysis challenges. The study populations for the multiple centers may differ geographically and ethnically, and considerable differences in case-control ascertainment and pedigree structures between studies are likely. PedGenie, available as an analysis option in Genie 2.4 software, requires Java 1.6 for execution and extends the functionality previously available in PedGenie 1.2 [1] by incorporating meta statistics for combined analysis of multi-study resources, along with Monte Carlo significance testing which allows for a mixture of pedigree members and independent individuals [2]. PedGenie offers a novel, structured approach that allows valid meta association testing, where each study considered can be a mixture of family-based individuals (including pedigrees of arbitrary size and complexity) and singletons, not simply pooling of data across studies. Briefly, study