Contributions
Abstract: 170
Type: Oral
Abstract Category: Pathology and pathogenesis of MS - Genetics /Epigenetics and Pharmacogenetics
Multiple Sclerosis (MS) exhibits variable prevalence across populations, with European populations having a higher prevalence than either Hispanic or African. In addition to environmental influences, this observation could be partly due to a greater genetic risk in European populations. Genetic association studies in individuals of European descent have identified 110 established MS risk variants in 103 discrete loci outside of the Major Histocompatibility Complex. Our goal was both to characterize the local ancestry across these loci in genetically admixed samples from Hispanic and African American populations as well as to use local ancestry to identify novel loci which may be related to MS. Data from the Illumina ImmunoChip were available on Hispanic (190 cases and 46 controls) and African American (973 cases and 257 controls) samples. Local ancestry was computed across 125,830 variants (inclusive of 186 fine-mapping regions for both MS and non-MS loci) with RFMix, after first phasing the haplotypes using Beagle. Reference data from 1000 Genomes (400 from each of the African, Asian, and European populations) were used for both phasing and local ancestry calculation. Separately in Hispanic and African American samples, logistic regression was used to test for the association of both the number of European haplotypes and the number of Asian haplotypes observed at each position, after controlling for global European and Asian ancestry. Global ancestry was computed by averaging local ancestry estimates across ~40,000 independent variants outside of known MS loci. In both Hispanic and African American populations we find evidence (p≤0.01) of European admixture difference across known MS loci between cases and controls. In African Americans, at a threshold of p≤0.001, in addition to the established CD58 locus, which was previously identified in an admixture scan of African Americans; we saw evidence for a decreased number of European haplotypes in cases at the NOD2 locus (a well-known Crohn"s disease locus). In Hispanics, at a threshold of p≤0.001, we saw an increase of both European and Asian haplotypes at the NXPE1 locus (an established ulcerative colitis locus). The increase in Asian haplotypes among cases was also observed in African Americans (p≤0.05). These results highlight the value of local ancestry evaluation in admixed populations to better characterize known risk loci and to potentially identify novel MS loci.
Disclosure: The authors have nothing to disclose.
Abstract: 170
Type: Oral
Abstract Category: Pathology and pathogenesis of MS - Genetics /Epigenetics and Pharmacogenetics
Multiple Sclerosis (MS) exhibits variable prevalence across populations, with European populations having a higher prevalence than either Hispanic or African. In addition to environmental influences, this observation could be partly due to a greater genetic risk in European populations. Genetic association studies in individuals of European descent have identified 110 established MS risk variants in 103 discrete loci outside of the Major Histocompatibility Complex. Our goal was both to characterize the local ancestry across these loci in genetically admixed samples from Hispanic and African American populations as well as to use local ancestry to identify novel loci which may be related to MS. Data from the Illumina ImmunoChip were available on Hispanic (190 cases and 46 controls) and African American (973 cases and 257 controls) samples. Local ancestry was computed across 125,830 variants (inclusive of 186 fine-mapping regions for both MS and non-MS loci) with RFMix, after first phasing the haplotypes using Beagle. Reference data from 1000 Genomes (400 from each of the African, Asian, and European populations) were used for both phasing and local ancestry calculation. Separately in Hispanic and African American samples, logistic regression was used to test for the association of both the number of European haplotypes and the number of Asian haplotypes observed at each position, after controlling for global European and Asian ancestry. Global ancestry was computed by averaging local ancestry estimates across ~40,000 independent variants outside of known MS loci. In both Hispanic and African American populations we find evidence (p≤0.01) of European admixture difference across known MS loci between cases and controls. In African Americans, at a threshold of p≤0.001, in addition to the established CD58 locus, which was previously identified in an admixture scan of African Americans; we saw evidence for a decreased number of European haplotypes in cases at the NOD2 locus (a well-known Crohn"s disease locus). In Hispanics, at a threshold of p≤0.001, we saw an increase of both European and Asian haplotypes at the NXPE1 locus (an established ulcerative colitis locus). The increase in Asian haplotypes among cases was also observed in African Americans (p≤0.05). These results highlight the value of local ancestry evaluation in admixed populations to better characterize known risk loci and to potentially identify novel MS loci.
Disclosure: The authors have nothing to disclose.