Non-reference haplotype frequencies (NHF) were calculated for each haplotype as the number of accessions with a haplotype deviating from the respective reference haplotype (FT1a_H0, FT1b_H0, FL1a_H0, FL1b_H0 or BTC1_H0) divided by the total number of accessions screened per B. vulgaris form.
Tests for haplotype frequency difference between the cases and controls were calculated for the haplotype blocks harboring the SNPs having P of <0.05 for allele association using the Haploview program.
REHH values were calculated for core haplotypes in all haplotype blocks for the relelvant chromosome.
The fixation indexes (FST) for pairwise comparisons between the risk haplotypes for continental populations were calculated for MYH9 haplotypes; FST ranged from 0.27 0.40 for Africa compared to other continental populations, possibly due to selection.
To infer population structures and historical processes resulting in the observed genetic distribution, a statistical parsimonious network was calculated for mtDNA haplotypes using the software TCS version 1.21 [32] at a 95% confidence limit.
Haplotype frequencies were calculated for 5-loci haplotypes and for combinations of HLA-B, C, DRB1, and DQB1 loci haplotypes containing the HLA-C*04:01 allele (Table 5 and Supplementary Table 2).
LTBP4 haplotypes were phased with PLINK, 20 and median ages at LoA were calculated for all observed haplotype configurations, in all those patients for whom haplotypes could be phased with at least 90% probability.
The haplo.glm function estimates haplotype frequencies with the EM algorithm and calculates for each haplotype linear regression coefficient and p-value, comparing each haplotype to a base haplotype, defined as the most common haplotype.
These networks did not show striking geographical structure, so we calculated, for each Korean haplotype, the distance to the closest northern and southern haplotype.
Haplotype distributions in different ethnic subgroups, along with median age at LoA calculated for all observed haplotype configurations (with standard errors [SEs] and 95% confidence intervals), are shown in detail in Supplementary Table 1.
