The approaches of population genetics and related disciplines are a powerful addition to the arsenal of weapons available for elucidating genetic influences on human diseases. The objective of this project is to distill existing and new theory in population genetics into strategies and algorithms applicable to the disease gene discovery effort of the LGD. We are applying these methods to studies of AIDS, prostate cancer, hypertension, kidney disease, and viral hepatitis. A principal focus has been the development of the mapping by admixture linkage disequilibrium (MALD) approach to mapping disease genes, whose feasibility we previously demonstrated by calculations and simulation studies. We have completed a proof-of principle study using the FY (duffy) locus as a surrogate for a disease gene. A test of 26 short tandem repeat (STR) loci in an 80 cM interval around FY showed linkage disequilibrium, detetectible by a contingency table analysis of different FY groups, for markers as far as 35 cM from FY, with all STR markers within 5 cM showing significant linkage disequilibrium. Our earlier MALD feasiblity studies have led to a full scale study of prostate cancer in African Americans. This study employs several hundred STR markers spaced at approximately 10 cM intervals across the genome, as suggested by the feasibility studies. Recognizing that in non-ideal circumstances (e.g., a disease etiology involving several low penetrance disease genes) the MALD signal for disease association may be weak, we have continued to refine methods for detecting MALD associations. The use of STR markers presents a specific mathematical challenge since many of the calculational methods originally developed for MALD assumed diallelic loci, while the STRs may have twelve or more alleles. We have developed a new method specific for multiallelic loci which uses allele frequency differences between the ancestral groups (with the African allele frequencies in this case being extrapolated from African American frequencies) to estimate ancestry in disease and control groups. This approach has been tested on markers around the FY (Duffy) locus, where it is able to detect linkage with FY with greater sensitivity than the contingency table analysis. Gene Mapping and AIDS - admixture, disease association, Evolution, linkage disequilibrium, Major histocompatibility Complex, microsatellites, recombination, - Human Tissues, Fluids, Cells, etc.
| Smith, Michael W; Patterson, Nick; Lautenberger, James A et al. (2004) A high-density admixture map for disease gene discovery in african americans. Am J Hum Genet 74:1001-13 |
