Genomic-Assisted Prediction of Genetic Value With Semiparametric Procedures

TLDR

Standard parametric quantitative genetic methods fail to handle the vast interactions among hundreds of thousands of markers and violate key variance‑decomposition assumptions in both artificial and natural populations. The authors propose semiparametric procedures that jointly use phenotypic and genomic data to predict total genetic value, arguing that nonparametric methods are preferable under these circumstances. They embed kernel regression and reproducing‑kernel‑Hilbert‑space regression into mixed‑effects linear models to manage massive SNP information, provide inferential procedures and extensions, and show that implementations can be achieved by modifying existing animal‑breeding software. An illustrative example demonstrates the potential of the methodology.

Abstract

Semiparametric procedures for prediction of total genetic value for quantitative traits, which make use of phenotypic and genomic data simultaneously, are presented. The methods focus on the treatment of massive information provided by, e.g., single-nucleotide polymorphisms. It is argued that standard parametric methods for quantitative genetic analysis cannot handle the multiplicity of potential interactions arising in models with, e.g., hundreds of thousands of markers, and that most of the assumptions required for an orthogonal decomposition of variance are violated in artificial and natural populations. This makes nonparametric procedures attractive. Kernel regression and reproducing kernel Hilbert spaces regression procedures are embedded into standard mixed-effects linear models, retaining additive genetic effects under multivariate normality for operational reasons. Inferential procedures are presented, and some extensions are suggested. An example is presented, illustrating the potential of the methodology. Implementations can be carried out after modification of standard software developed by animal breeders for likelihood-based or Bayesian analysis.