Supervised independent component analysis as an alternative method for genomic selection in pigs

Abstract

The objective of this work was to evaluate the efficiency of the supervised independent component regression (SICR) method for the estimation of genomic values and the SNP marker effects for boar taint and carcass traits in pigs. The methods were evaluated via the agreement between the predicted genetic values and the corrected phenotypes observed by cross-validation. These values were also compared with other methods generally used for the same purposes, such as RR-BLUP, SPCR, SPLS, ICR, PCR and PLS. The SICR method was found to have the most accurate prediction values.

Comparison of dimensionality reduction methods to predict genomic breeding values for carcass traits in pigs

“…Azevedo et al (2013b) compared the PLS, PCR, and ICR methods and reported results of lower accuracy, from 0.01 to 0.51, compared to those found in this study. Azevedo et al (2014) compared the same methods of dimensional reduction beyond their supervised approaches (selection of covariates), showing different results to those reported in this study, where PLS had low performance compared to PCR and ICR.…”

“…However, the advantage of ICR for GWS compared to PCR is the fact that it considers complete independence of the components, guaranteeing the absence of both linear and nonlinear relationships between latent variables (Azevedo et al, 2013(Azevedo et al, , 2014. Similar results observed between the PCR and the ICR methods are possibly due to the structure of association of the variables evaluated in this paper, which have linear behavior.…”

“…The use of dimensional reduction methods has only been reported by Azevedo et al (2013a,b) and Azevedo et al (2014) for genomic selection in carcass traits in the same population of pigs, but using a different criterion for choosing the number of components. Azevedo et al (2013a) studied the performance of multivariate and univariate approaches of the PLS method, but the results disagreed with the values obtained in our study, where MPLS outperformed UPLS considerably.…”

Genomic selection in pigs: state of the art and perspectives

“…These two components were considered both in GWAS studies (be e.g. in Duijvesteijn et al 2010) and more recently in GEBV prediction analyses (Azevedo et al 2014;de Campos et al 2015). In detail, de Campos et al (2015) reported GEBV accuracy values of 0.65 for androstenone and 0.58 for skatole levels.…”

The optimal number of partial least squares components in genomic selection for pork pH

“…According to AZEVEDO et al (2014), the advantage of PLS over PCR is due to the fact that the former method takes account of variable response (phenotypic observations) when processing components' composition, whereas the PCR method considers only the covariates themselves. Poor performance of the OLS method may be due to the lack statistical features to control the problems of multicollinearity, thereby leading to overfitting problems (SILVEIRA et al, 2014) when compared to the PLS and PCR methods.…”