Bayesian genomic-enabled prediction as an inverse problem.
Bottom Line: Many prediction models address the collinearity that arises when the number (p) of molecular markers (e.g. single-nucleotide polymorphisms) is larger than the sample size (n).Here we propose four Bayesian approaches to the problem based on commonly used data reduction methods.Because shrinkage of estimates is affected by the prior variance of transformed effects, we propose four structures of the prior variance as a way of potentially increasing the prediction accuracy of the models fitted.
Affiliation: Colegio de Posgraduados, 56230, Montecillo, Texcoco, Edo. de México.Show MeSH
Related in: MedlinePlus
Mentions: Figure 2, A and B depicts the prior variance for models BIRR, BAI, BIR1, and BIR2 for trait-environment combination MFL-WW. In Figure 2A, the prior variance of for BIRR is represented by a solid line, whereas used in BAI are scattered dots, each representing an individual. It is interesting to note that, for MFL-WW, most of the values for BAI, BIR1, and BIR2 were smaller than those for BIRR, represented by a solid line. This indicates that BAI, BIR1, and BIR2 cause more shrinkage. Figure 2B depicts the decay of singular values for BIR2 (dashed line) and BIR1 (solid line), both mimicking the current decay of singular values shown in Figure 1, A and B. The decay of BIR1 reflects the polynomial function , but smoothed by the parameter h, as indicated in (18), with less shrinkage for the first singular values and increasing shrinkage toward the later singular values.
Affiliation: Colegio de Posgraduados, 56230, Montecillo, Texcoco, Edo. de México.