Purchase this article with an account.
S. Schmidt, M. A. Hauser, R. R. Allingham, P. Gallins, W. K. Scott, A. Agarwal, J. L. Haines, E. A. Postel, M. A. Pericak-Vance; Covariate-Specific Linkage Peaks From a High Density SNP Genome Screen for Age- Related Macular Degeneration. Invest. Ophthalmol. Vis. Sci. 2007;48(13):2098.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
AMD is a common complex disorder with both genetic and environmental contributions. With two major susceptibility genes already identified, it is likely that additional genes have smaller effects, confer subtype-specific risk, and/or act primarily in the presence of environmental risk factors. We report results of incorporating environmental and clinical covariates into an SNP-based genome-wide linkage screen for AMD.
The Illumina Bead Station platform (Linkage Panel IV) was used to genotype 339 individuals from 115 multiplex AMD families. After excluding markers that did not meet quality control criteria, 5253 SNPs were analyzed with standard linkage methods. To incorporate continuous environmental and clinical covariates, ordered subset analysis (OSA) and quantitative trait locus (QTL) analysis were applied to the following variables: age at exam, body mass index (BMI), intraocular pressure (IOP) and pack-years of cigarette smoking.
We replicated previously reported OSA results for chromosome 14q13 in our current dataset, in which the multipoint lod score for a subset of 61 families with below-average IOP (≤15.5) values increased from 1.6 to 4.1 at 33 cM (p=0.02). Chromosome 12q23, which is one of the top nine susceptibility regions reported by a recent meta analysis, may harbor a locus that influences the AMD risk specifically in cigarette smokers: Analyzing pack-years of smoking as the trait in a QTL analysis yielded a lod score of 3.6 at 117 cM. On chromosome 6p12, analyzing BMI as the trait gave a lod score of 3.2 at 73 cM (~47.4 Mb), and using age at exam as the OSA covariate gave a peak score of 2.5 at 75 cM for 63 families with average age at exam ≤80 years (p=0.04). BMI and age at exam are significantly correlated in our dataset (r=-0.225, p<0.0001). It is currently unclear whether this finding is partially explained by the recently implicated factor B and complement component 2 genes located at ~37.0 Mb.
The incorporation of clinical and environmental covariates into our genome screen analysis has narrowed down previously reported linkage regions, for which the underlying susceptibility genes have yet to be identified.
This PDF is available to Subscribers Only