Evidence for single nucleotide polymorphisms having a nonadditive influence within the apolipoprotein E gene.

• Main Author: Hamon, Sara Christina
• Other Authors: Sing, Charles F.
• Format: Thesis
• Language: English
• Published: 2003
• Subjects: Apolipoprotein E; Cardiovascular Disease; Epistatic; Evidence; Gene; Having; Influence; Nonadditive; Single-nucleotide Polymorphisms; karelia*
• Online Access: https://hdl.handle.net/2027.42/123851; http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&res_dat=xri:pqm&rft_dat=xri:pqdiss:3106071

Current research to determine the influence of a SNP on levels of a quantitative trait consider only the additive or marginal contribution of a SNP. This strategy for selecting SNPs completely ignores the genetic background of an individual and discounts the possibility that a SNP may only have an influence in the presence of a second SNP. One method to determine if a pair of SNPs is contributing to quantitative trait levels in a non-additive manner is the overparameterized model. This thesis relates the overparameterized model to the traditional models of epistasis developed by Fisher (1918), Cockerham (1954), Kempthorne (1954) and Cheverud and Routman (1995). We analyzed 20 single nucleotide polymorphisms (SNPs) within the apolipoprotein E ( APOE ) gene to identify pairs of SNPs that interact in a non-additive manner to influence mean levels of the ApoE protein in the blood. An overparameterized general linear model of two-SNP genotype means was applied to all pairs of 16 SNPs segregating in a sample of 483 unrelated female and a sample of 219 unrelated male African Americans from Jackson, Miss, of 14 SNPs segregating in a sample of 162 unrelated female and a sample of 124 unrelated male Europeans from North Karelia, Finland and of 13 SNPs segregating in a sample of 456 unrelated female and a sample of 398 unrelated male European Americans from Rochester, Minn. We established that there was statistically significant evidence for non-additivity within one of the six gender-population strata, Rochester males. In this sample we observed nine pairs of SNPs with evidence of non-additivity at alpha = 0.05 of statistical significance when approximately three were expected by chance. Five of the nine pairs involved three SNPs (560, 624 and 1163) that did not have statistically significant influence when considered separately in a single site analysis. For the four of the nine pairs of SNPs where the same comparisons of two-SNP genotypes were possible, we evaluated whether the non-additive effects in Rochester males ...