Extending Admixture Mapping to Nuclear Pedigrees: Application to Sarcoidosis. Issue 3 (31st January 2013)

• Record Type: Journal Article
• Authors: McKeigue, Paul M.; Colombo, Marco; Agakov, Felix; Datta, Indrani; Levin, Albert; Favro, David; Gray‐Montgomery, Courtney; Iannuzzi, Michael C.; Rybicki, Benjamin A.
• Is Part Of: Genetic epidemiology. Volume 37:Issue 3(2013)
• Page Start: 256
• Page End: 266
• Publication Date: 2013-01-31
• Subjects: Genetic epidemiology -- Periodicals; Heredity -- Periodicals; Medical geography -- Periodicals; 614
• DOI: 10.1002/gepi.21710
• Languages: English
• ISSNs: 0741-0395

Abstract

<abstract abstract-type="main"> <title> <x xml:space="preserve">Abstract</x> </title> <p>We describe statistical methods that extend the application of admixture mapping from unrelated individuals to nuclear pedigrees, allowing existing pedigree‐based collections to be fully exploited. Computational challenges have been overcome by developing a fast algorithm that exploits the factorial structure of the underlying model of ancestry transitions. This has been implemented as an extension of the program ADMIXMAP. We demonstrate the application of the method to a study of sarcoidosis in African Americans that has previously been analyzed only as an admixture mapping study restricted to unrelated individuals. Although the ancestry signals detected in this pedigree analysis are generally similar to those detected in the earlier analysis of unrelated cases, we are able to extract more information and this yields a much sharper exclusion map; using the classical criterion of an LOD score of minus 2, the pedigree analysis is able to exclude a risk ratio of 2 or more associated with African ancestry over 96% of the genome, compared with only 83% in the earlier analysis of unrelated individuals only. Although the pedigree extension of ADMIXMAP can use ancestry‐informative markers only at relatively low density, it can use imputed ancestry states from programs such as WINPOP or HAPMIX that use dense SNP marker genotypes for admixture mapping. This extends both the efficiency and the range of application of this powerful gene mapping method.</p> </abstract>