The Irish DNA Atlas: Revealing Fine-Scale Population Structure and History within Ireland
By Edmund Gilbert, Seamus O’Reilly, Michael Merrigan, Darren McGettigan, Anne M. Molloy, Lawrence C. Brody, Walter Bodmer, Katarzyna Hutnik, Sean Ennis, Daniel J. Lawson, James F. Wilson and Gianpiero L. Cavalle
Nature: Scientific Reports 7:17199 (2017)
Introduction: Located off the North-Western seaboard of Europe, Ireland’s geographic situation is conducive to genetic homogeneity and isolation. Indeed, several traits are found to be at high frequencies within the Irish, compared to the mainland European populations, including; cystic fibrosis, lactase persistence, coeliac disease, galactosaemia, and multiple sclerosis. Studies of ancient Irish genomes suggest that the modern Irish genetic landscape was established about 3,500 years ago in the Irish Bronze Age. There have since been a number of significant historical migrations into Ireland; the Norse Vikings in the late first millennium, the Norman invasion of the 12th century, and the Plantations of the 16th and 17th centuries. The impact of these migrations on the modern Irish genome is largely unknown.
Previous attempts to describe genetic structure in the Irish population primarily employed uniparental markers. Y-chromosome and mitochondrial haplotypes that are common within Ireland show genetic continuity with those observed in other western European populations. Within the Irish population, there appears to be significant geographic structure in Y-chromosome haplotypes. There exists a general east-west cline of declining Y-chromosome haplotype diversity, evidence of hegemony in the north-west of Ireland, and in Munster, two haplotypes associated with the north and south of that province, respectively. Furthermore, analysis of putatively Norse-origin surnames suggests little Norse Y-chromosome introgression within Ireland.
The first autosomal evidence for population structure within Ireland resulted from analysis of ABO and Rhesus blood group frequencies. Despite the limited resolution, this early work showed a general east-west cline in blood groups, mirroring the Y-chromosome results that would follow. Recent work with dense, genome-wide SNP data has established that Irish haplotype diversity is reduced relative to other European populations, and both Runs of Homozygosity (ROH) and Linkage Disequilibrium (LD) have been found to be slightly elevated in Irish genomes. Previous attempts have been unable to detect population structure within Ireland, only observing that the population appears to be divergent from the neighbouring Scottish and English populations. A limitation of these studies was a lack of geocoded data sampling, preventing any analysis of regions within these populations. Recent work has demonstrated fine-scale structure within the British Isles. However the extent of such structure across the island of Ireland is unknown, and such a description would facilitate disease genemapping within Ireland, as well as complementing similar efforts involving populations with Irish ancestry.
In the current study we report on the assembly of a cohort of individuals with extended ancestry from specific regions in Ireland. Using dense genome-wide SNP genotype data we have performed a number of analyses to investigate population structure and extent of admixture within Ireland. We first utilize both fineStructure to investigate the extent of fine-scale population ‘clusters’, and Estimated Effective Migration Surfaces analysis to visualize regions of low or high genetic migration within Ireland and Britain. Second, we apply regression-based admixture modelling analysis and admixture event detecting software (Globetrotter) to elucidate the extent of admixture in Ireland.