Nonparametric Linkage Analysis for Complex Genetic Diseases with Linked Loci

During the past two decade, linkage analysis has been phenomenally successful in localizing Mendelian disease genes. Now, human genetics moves toward identification of genes contributing to the susceptibility to common and non-Mendelian diseases, such as hypertension, diabetes, heat disease, familial cancer, and so on. In parametric linkage analysis, mode of inheritance of the disease susceptibility loci must be specified, but for non-Mendelian traits, it is too daunting. In such case, nonparametric linkage analysis (NLA) methods are preferred, because those require no assumptions about mode of inheritance of traits. Most complex disease are governed by multiple agent loci, but the analysis is often carried out under the implicit assumption that those are governed by a single locus, for practical reasons. This may cause misestimation of recombination fractions [3]. Recently, several nonparametric statistical tests have been proposed for the simultaneous detection of multiple loci involved in complex diseases, and it is known that those tests can reach higher power than single locus tests [1, 4]. But assumptions of those multi-gene methods are limiting. Especially, those methods do not allow linkage between agent loci. Here we propose new algorithm of nonparametric linkage analysis that allow the linkage between agent loci and arbitrary size of pedigrees.