Purchase this article with an account.
John Paul SanGiovanni; A Pathway-based Genome-wide Analysis Yields Multi-locus AMD Associations in Genes Encoding JNK/MAPK Pathway Elements, Platelet Activation Triggers, and Targets of miR34 and HDAC1. Invest. Ophthalmol. Vis. Sci. 2013;54(15):4107.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
We applied a pathway-based genome-wide analysis to determine whether DNA sequence variants within groups of functionally related genes are, in aggregate, more strongly associated with advanced AMD (AAMD) than expected by chance alone.
We used age-, sex-, and smoking-adjusted meta-regression to compute combined estimates of gene-AMD relationships from 3 independent cohorts examined in large-scale genotyping projects on the molecular genetics of AMD (1177 people with AAMD and 1024 of their AMD-free peers who were > 65 years-of-age). SNPs spanning independent genomic intervals and significantly related with AAMD at P<0.002 were tested with exact methods for enriched association signals within functionally-defined gene sets using a validated 2-phase genomic permutation procedure applying 5000 and 1000 iterations. The multi-locus analytic method uses positional clustering of SNPs to account for factors (varying gene size, SNP density, linkage disequilibrium) that otherwise may lead to non-differential misclassification and inflation in measures of association.
AAMD-associated SNPs were enriched in 8 independent genomic regions of 7 genes in the JNK/MAPK pathway, yielding an exact P-value of 0.002 for the likelihood of a chance finding. Twelve independent genomic regions in 10 genes encoding platelet activation triggers carried AAMD-associated sequence variants; the exact likelihood of observing a random clustering of findings within this functionally related gene set was 0.002. Seventeen genes containing a specific microRNA (miR34) target carried 23 independent regions with AAMD-associated SNPs (exact P = 0.004). Thirty-four independent genomic regions in 34 genes that are up-regulated by histone deacetylase 1 (HDAC1) contain AAMD-associated variants (exact P = 0.007).
A multi-locus genome-wide analysis yielded associations of AAMD with aggregates of functionally related genes encoding 10 platelet activation triggers, 7 elements of the JNK/MAP signaling pathway, and 23 and 34 respective targets of the epigenetic regulatory elements miR34 and HDAC1. Our findings may be used to identify molecular probes for drug testing and determination of metabolic fate.
This PDF is available to Subscribers Only