Skip to main content
Figure 3 | BMC Bioinformatics

Figure 3

From: Genomic selection and complex trait prediction using a fast EM algorithm applied to genome-wide markers

Figure 3

Graphical illustration of how a posterior probability is calculated for a SNP. Graphs of the mixture prior p(g j ), conditional likelihood h(g j |G j , σ2) and conditional posterior distribution p(g j |g -j, y) as given in equations (A1) and (A2) of Additional file 1 for γ = 0.05 , λ = 10 , σ e 2 = 1 and n = 500. The Dirac Delta function is replaced by a DE with λ s = 1000 i.e. a Spike at 0. The posterior probability γ j of SNP j being in LD with QTL is calculated from equation (A3) by numerical integration. Figures A and B show the distributions when G j is 0.19 and 0.11 respectively, where G j is the conditional maximum likelihood estimate of g j .

Back to article page