PEDIGREE MODELS FOR COMPLEX HUMAN TRAITS INVOLVING THE MITOCHONDRIAL GENOME

Recent biochemical and molecular-genetic discoveries concerning variat ions in human mtDNA have suggested a role for mtDNA mutations in a num ber of human traits and disorders. Although the importance of these di scoveries cannot be emphasized enough, the complex natures of mitochon drial biogenesis, mutant mtDNA phenotype expression, and the maternal inheritance pattern exhibited by mtDNA transmission make it difficult to develop models that can be used routinely in pedigree analyses to q uantify and test hypotheses about the role of mtDNA in the expression of a trait. In the present paper, we describe complexities inherent in mitochondrial biogenesis and genetic transmission and show how these complexities can be incorporated into appropriate mathematical models. We offer a variety of likelihood-based models which account for the c omplexities discussed. The derivation of our models is meant to stimul ate the construction of statistical tests for putative mtDNA contribut ion to a trait. Results of simulation studies which make use of the pr oposed models are described. The results of the simulation studies sug gest that, although pedigree models of mtDNA effects can be reliable, success in mapping chromosomal determinants of a trait does not preclu de the possibility that mtDNA determinants exist for the trait as well . Shortcomings inherent in the proposed models are described in an eff ort to expose areas in need of additional research.