Infinium platform

All CC lines were genotyped with high-density SNP markers using the MDA (620 K SNP markers), MUGA (7.7 K markers) and MegaMUGA (77 K markers) genotyping arrays based on the Illumina infinium platform. All SNPs with heterozygous or missing genotypes in the 8 CC founders or not common between the arrays were filtered out, leaving 170,935 SNPs. The SNPs were mapped onto build 37 of the mouse genome. A descent probability distribution was computed using the HAPPY HMM for each of the 170,935 SNPs intervals. The genotype status of the CC lines using the MUGA SNP was presented in CTC 2012 report [9 (link), 25 (link)].

F₃₄ animals had been genotyped on a custom SNP array on the Illumina Infinium platform (Cheng et al. 2010 (link); Parker et al. 2014 (link)), which yielded a set of 4,593 SNPs on autosomes and X chromosome that we refer to as ‘sparse SNPs’.

Genomic DNA of each accession was extracted from young leaf tissues for genotyping using a modified CTAB method (Zhang and Stewart, 2000). A CottonSNP63K array containing 63 058 SNPs (Hulse‐Kemp et al., 2015), which was recently developed by an international cotton SNP consortium, was applied to genotype the 719 accessions using the Illumina Infinium platform according to the manufacturer's protocol. All the SNP data were clustered and selectively analysed by Illumina GenomeStudio genotyping software. The SNP data set was further filtered with a calling rate < 0.85 and MAF < 0.05. For the physical localization of SNP markers, the probe sequences of the SNPs were used to perform a local BLAST (Altschul et al., 1990) search against the G. hirsutum TM‐1 reference genome (Zhang et al., 2015b). SNPs that could not be assigned to a G. hirsutum chromosome were excluded from further analysis.

DNA was extracted and purified from tail biopsies taken from 6-wk-old mice using QIAGEN DNeasy kit according to the manufacturer’s instructions. Genotyping was performed using the Mega Mouse Universal Genotyping Array (MegaMUGA) by GeneSeek (Neogen, Lansing, MI) (Welsh et al. 2012 (link)). The MegaMUGA array is built on the Illumina Infinium platform and contains 77,800 SNP markers that are distributed throughout the genome at an average spacing of 33 kb. For the mapping, genomes were reconstructed based on the X and Y allele intensities from the array and founder haplotypes were reconstructed using a hidden Markov model. The reconstructed haplotypes were then used to perform linkage mapping.

The WTCCC methods for genotyping and quality control have been described previously [12 (link)–14 (link), 16 (link), 18 (link), 19 ]. Genotyping was performed at the Wellcome Trust Sanger Institute on the Illumina Infinium platform and case samples were genotyped using a customized Human660-Quad chip. Common controls were genotyped on a second customized Human1M-Duo chip (utilizing the same probes). This provided data on 441,547 autosomal SNPs scattered throughout the genome in both MS patients and control subjects after quality control. We used the the HIBAG method [40 (link)] to impute the identities of the five HLA alleles in the MHC region (A, C, B, DRB1 and DQB1). Imputation methods, apart from HIBAG, such as the so-called ‘SNP2HLA’ method [41 (link)], have also been proposed. However, in a comparative study in European Americans, the HIBAG method had slightly better concordance rates and had slightly worse call rates compared to SNP2HLA [42 (link)]. Nevertheless, these differences were minimal, and both imputation methods seem equally good and the use of either is appropriate [42 (link)].

IGAS test cohort samples were genotyped using the Illumina Immunochip array on the Illumina Infinium platform as previously described [5 (link)]. UKB replication cohort controls were genotyped on the UKB Affymetrix Axiom array. Identity by decent was calculated using the—genome command in PLINK [75 (link)], with exclusion of one individual from each pair with a PI_HAT score >0.05. Principal components (PCs) for ethnicity confirmation and population stratification correction were calculated for test and replication cohorts independently based on 20,783 and 12,485 autosomal SNPs respectively outside of long-range LD regions [76 (link)]. Homogeneity of ethnic background was confirmed by visualisation of the first two PCs with exclusion of individuals falling beyond plus or minus three standard deviations from the mean of the European sample cluster. PCs were recalculated for the filtered European test and replication cohorts (S1 Fig shows stratification of the European cluster along the first two principal components) and the first ten principal components fitted as covariates in all regression models.

Genomic DNA was extracted from soybean seedling leaves according to the methods used by Kisha et al. [31 (link)], and DNA quality was detected by 1% agarose gel electrophoresis and a spectrophotometer. A genome-wide genotyping array containing 158,327 SNPs was applied to genotype the 410 accessions using the Illumina Infinium platform, according to the manufacturer’s protocol (Illumina) [32 (link),33 (link)]. All SNP genotype data were treated with raw data normalization, clustering, and genotype calling using Illumina Genome Studio Genotyping Module (Illumina). The SNPs with a minor allele frequency (MAF) < 0.05 and missing rates < 0.25 were removed to avoid problems of spurious LD and false positive associations. Finally, 117,811 high-quality SNPs were used for GWAS analysis. The SNPs were distributed relatively evenly across the 20 soybean chromosomes (Figure S1).