Computational protocol: Genome-Wide Association Analysis for Blood Lipid Traits Measured in Three Pig Populations Reveals a Substantial Level of Genetic Heterogeneity

Similar protocols

Protocol publication

[…] Heritability of each trait was estimated by using –lmm procedure of GEMMA based on genomic relationship matrix []. A mixed model was used to analyze the associations of the eligible SNPs with blood lipid traits: Y = Xb + Sα+ Zμ+ e, where b is the fixed effects that included sex and batch (the batch was 22, 11 and 4 for DLY, Erhualian and Laiwu, respectively); X is the incidence matrix of the fixed effects; α is the SNP substitution effect; S is the incidence matrix for α; μ is the vector of random additive genetic effects that follow the distribution N (0, Gσ2), where G is the kinship matrix derived from SNP markers [, ] and σ2 is the additive variance; Z is the identity matrix for μ; e is the residual error. The mmscore function of GenABEL was used to estimate the significance of associations between SNP markers and target traits []. A bonferroni correction was applied to determine the genome-wide (P < 0.05/SNP number) and suggestive (P < 1/SNP number) significance thresholds. The meta-analysis of GWAS was performed on the five populations of F2, Sutai, DLY, Laiwu and Erhualian pigs by employing METAL []. In brief, for each marker, the same reference allele was selected for all tested populations and a Z-score for evidence of association was calculated. The Z-statistics summarized the magnitude and the direction of allelic effect relative to the reference allele. An overall Z-score and p-value were then calculated from a weighted sum of the individual statistics. Weights were proportional to the square-root of the number of individuals examined in each population. The porcine genome assembly 10.2 was retrieved to characterize functionally plausible candidate genes (http://www.ensembl.org/Sus_scrofa/Location/Genome).Population stratification affects the validity of genome-wide association study []. Population stratification was corrected by fitting the covariance among individuals that was inferred from high density SNP data. Moreover, genomic control (GC) was used to correct the effect of stratification (λ) that was estimated from the null test statistics (under the null hypothesis of no SNP associated with the trait) []. Here, we evaluated population stratification by examining the distribution of test statistics in a quantile-quantile (Q-Q) plot []. The Q-Q plots were constructed with R software. […]

Pipeline specifications

Software tools GEMMA, GenABEL
Application GWAS
Organisms Sus scrofa, Homo sapiens
Diseases Cardiovascular Diseases, Myocardial Infarction