Additional file 2 of Sequence-based genome-wide association study of individual milk mid-infrared wavenumbers in mixed-breed dairy cattle

Additional file 2: Table S2. Peak variants of 27 protein-sequence association effects classified as moderately significant for FT-MIR wavenumber phenotypes. Moderately significant effects are those for which the −log10(p-value) of the effect was greater than 1x the Bonferroni threshold of −log10(6.2e-13) and the correlation between the tag variant and the protein-sequence variant was higher than 0.9, but the effect did not meet the criteria of a highly significant effect (see Table 1). Effects where the locus has been identified as highly significant based on the LD with one or more other genes (and is also present in Table 1) are shaded yellow. No. of hits is the number of wavenumbers for which the variant was selected as the representative (most significant) tag variant for a peak. Iterations are defined relative to the base GWAS, with the base GWAS represented as iteration 0. L = Low impact splice region variant; M = Moderate impact missense variant; H = High impact splice donor. Table S3. Minor allele frequencies and allele effects for WGS tag variants with a highly significant protein-sequence association effect in at least one FT-MIR wavenumber. Table S4. Association statistic profiles for 31 highly significant protein-sequence effects identified in FT-MIR wavenumber phenotypes. For each protein-sequence mutation, the proportion of phenotypic and genetic variance that it accounts for is shown for each of the 895 FT-MIR wavenumbers and three FT-MIR predicted milk composition phenotypes. The genetic variance for each phenotype is the SNP-based estimate evaluated by the Bolt-LMM software. Table S5. Chronological profiles across iterations for 17 highly significant protein-sequence association effects observed in FT-MIR wavenumbers, where the association is observed after fitting the top chromosomal variant(s) in previous GWAS iterations and/or the base GWAS as covariates. Iterations are defined relative to the base GWAS, with the base GWAS represented as iteration 0. P-values at previous iterations for the phenotype, and p-values for the corresponding top chromosomal SNP in that iteration are provided. Table S6. Peak variants of 14 protein-sequence association effects classified as moderately significant for FT-MIR predicted milk composition traits. Moderately significant effects are those where the −log10(p-value) of the effect was greater than 1x the Bonferroni threshold of −log10(6.2e-13) and the correlation between the tag variant and the protein-sequence variant was higher than 0.9, but the effect did not meet the criteria of a highly significant effect (see Table 2). Effects where the locus has been identified as highly significant based on the LD with one or more other genes (and is also present in Table 2) are shaded yellow. Iterations are defined relative to the base GWAS, with the base GWAS represented as iteration 0. FP = Fat %; LP = Lactose %; PP = Protein %; L = Low impact splice region variant; M = Moderate impact missense variant; H = High impact splice donor. Table S7. Minor allele frequencies and allele effects for WGS tag variants with a highly significant protein-sequence association effect in at least one FT-MIR predicted milk composition trait. FP = Fat %; LP = Lactose %; PP = Protein %. Table S8. Chronological profiles across iterations for highly significant protein-sequence association effects observed in FT-MIR predicted milk composition traits, where the association is observed after fitting the top chromosomal variant(s) in previous GWAS iterations and/or the base GWAS as covariates. Iterations are defined relative to the base GWAS, with the base GWAS represented as iteration 0. P-values at previous iterations for the phenotype, and p-values for the corresponding top chromosomal SNP in that iteration have been provided. FP = Fat %; LP = Lactose %; PP = Protein %.