Improving the Efficiency of Genomic Selection

TL;DR

This paper explores feature selection via Markov blankets and kinship adjustments to improve genomic selection efficiency, demonstrating that these methods maintain or enhance predictive accuracy while reducing costs.

Contribution

It introduces a theoretically-sound feature selection method and a novel kinship measure for GBLUP, both improving efficiency in genomic prediction models.

Findings

Feature selection maintains or improves predictive power.

LD-adjusted kinships perform similarly to elastic net with feature selection.

Both methods outperform traditional approaches in real-world datasets.

Abstract

We investigate two approaches to increase the efficiency of phenotypic prediction from genome-wide markers, which is a key step for genomic selection (GS) in plant and animal breeding. The first approach is feature selection based on Markov blankets, which provide a theoretically-sound framework for identifying non-informative markers. Fitting GS models using only the informative markers results in simpler models, which may allow cost savings from reduced genotyping. We show that this is accompanied by no loss, and possibly a small gain, in predictive power for four GS models: partial least squares (PLS), ridge regression, LASSO and elastic net. The second approach is the choice of kinship coefficients for genomic best linear unbiased prediction (GBLUP). We compare kinships based on different combinations of centring and scaling of marker genotypes, and a newly proposed kinship measure that adjusts for linkage disequilibrium (LD). We illustrate the use of both approaches and examine their performances using three real-world data sets from plant and animal genetics. We find that elastic net with feature selection and GBLUP using LD-adjusted kinships performed similarly well, and were the best-performing methods in our study.