Figure S6From: Inference of ancestry: constructing hierarchical reference populations and assigning unknown individualsDistribution of mis-assigned unknown individuals. Individuals are categorised by degree of mis-assignment (y-axis), and on the continuous variable x-axis quantified by proportional difference in average gene diversity between mis-assigned and original clusters. Degree of mis-assignment was quantified by the number of tiers required to be regressed in order for the two clusters involved to converge.Back to article page