|
A. Specific Aims
Meiotic linkage maps are the foundation of both linkage and linkage disequilibrium studies for mapping disease genes. Despite the importance of precise maps, existing genome-wide linkage maps were built using only a small collection of pedigrees, and so have wide confidence intervals surrounding estimates of map distance. These inaccuracies can contribute to misleading results, and may be one of the reasons that disappointingly few genes have been definitively identified that contribute to such complex diseases as asthma, cardiovascular disease, hypertension, hypercholesterolemia, diabetes, obesity, and cancer. We propose to build improved highly-precise linkage maps utilizing thousands of individuals who have previously been genotyped. In addition to creating precise maps for the scientific community, we also propose to use these genotype data to study how recombination may vary between ethnic groups. Our previous individual and combined work demonstrates our commitment to these areas and highlights our unique abilities to effectively pursue them.
Specific Aim 1: To collect, database, and filter genotypes that were previously determined by the Mammalian Genotyping Service.
We have requested permission to use genome-wide scan data from 40 PI's of studies that have been genotyped at the Mammalian Genotyping Service (MGS) at the Center for Medical Genetics in Marshfield, IL, which is funded by the National Heart, Lung, and Blood Institute (NHLBI). To date we have already received such permission from 28 investigators, and we anticipate obtaining permission from 12 additional investigators, whose total data include over 37,800 individuals from over 3,400 pedigrees. We will obtain these data from the individual investigators, combine them with other publicly available genotype data previously used to construct linkage maps (the Centre d'Etude du Polymorphism Humain (CEPH) pedigrees), create a database for local storage, and apply several filtering steps to remove unambiguous genotyping errors and pedigree structure errors.
Specific Aim 2: To construct linkage maps with very precise estimates of map distance and accurate marker orders.
The currently available genome-wide linkage maps were constructed using only eight CEPH pedigrees, which contain only 92 offspring for a total of 184 meioses. Using our expanded set of more than 3,400 MGS pedigrees plus the 8 CEPH pedigrees, we will validate marker orders and employ likelihood methods that model genotyping errors to produce very precise estimates of map distances, both for sex-averaged and sex-specific maps. These data will provide a unique opportunity to test the hypothesis that distributions of recombination events are the same across different ethnic sub-populations, an area in which very little prior work has been done. Therefore, genotypes from ethnic subgroups will be evaluated separately to test for group-specific distributions of recombination. If we find no evidence for group-specific recombination, then all genotypes will be combined together; alternatively novel sub-group specific maps will be created.
Our proposal to construct improved human linkage maps based on existing genotype data will provide a novel resource that will better facilitate disease gene studies by providing very precise estimates of map distance and marker order. Improved sex-specific map distances will be of particular value. In addition, detailed evaluation of whether recombination varies among ethnic groups will improve our limited understanding of how recombination is distributed in humans. We will make our new maps available to the scientific community via a user-friendly interactive web site, as well as incorporate much of the underlying data into our Web-based server (MAP-O-MAT) for automated linkage mapping of markers.
|