Including Phenotypic Causal Networks in Genome-Wide Association Studies Using Mixed Effects Structural Equation Models

Abstract

Network based statistical models accounting for putative causal relationships among multiple phenotypes can be used to infer single-nucleotide polymorphism (SNP) effect which transmitting through a given causal path in genome-wide association studies (GWAS). In GWAS with multiple phenotypes, reconstructing underlying causal structures among traits and SNPs using a single statistical framework is essential for understanding the entirety of genotype-phenotype maps. A structural equation model (SEM) can be used for such purposes. We applied SEM to GWAS (SEM-GWAS) in chickens, taking into account putative causal relationships among breast meat (BM), body weight (BW), hen-house production (HHP), and SNPs. We assessed the performance of SEM-GWAS by comparing the model results with those obtained from traditional multi-trait association analyses (MTM-GWAS). Three different putative causal path diagrams were inferred from highest posterior density (HPD) intervals of 0.75, 0.85, and 0.95 using the inductive causation algorithm. A positive path coefficient was estimated for BM → BW, and negative values were obtained for BM → HHP and BW → HHP in all implemented scenarios. Further, the application of SEM-GWAS enabled the decomposition of SNP effects into direct, indirect, and total effects, identifying whether a SNP effect is acting directly or indirectly on a given trait. In contrast, MTM-GWAS only captured overall genetic effects on traits, which is equivalent to combining the direct and indirect SNP effects from SEM-GWAS. Although MTM-GWAS and SEM-GWAS use the similar probabilistic models, we provide evidence that SEM-GWAS captures complex relationships in terms of causal meaning and mediation and delivers a more comprehensive understanding of SNP effects compared to MTM-GWAS. Our results showed that SEM-GWAS provides important insight regarding the mechanism by which identified SNPs control traits by partitioning them into direct, indirect, and total SNP effects.

Gut microbiome mediates host genomic effects on phenotypes: a case study with fat deposition in pigs

“…In a study on beef cattle performed with a different approach, Leal-Gutiérrez et al [27] found different genomic regions affecting meat quality (expressed as a latent variable) either directly or mediated by carcass quality. Similarly, again with a different approach, Momen et al [33] and Momen et al [34] identified several genomic markers with phenotype-mediated effects in chicken and rice, respectively.…”

A Multiple-Trait Bayesian Variable Selection Regression Method for Integrating Phenotypic Causal Networks in Genome-Wide Association Studies

“…Assume in equation (1) and follow assumptions in multi-trait BayesC , the SEM-Bayes model can be written as: Move from the right side to the left side of equation (3), and define , where is a identity matrix and is a matrix of structural coefficients based on the discerned causal structure, the model becomes: To guarantee that the structural coefficient is identifiable, we assume that the residuals for each trait of individual i are independent with each other, which means the residual covariance matrix is diagonal ( Wu et al 2010 ; Momen et al 2018) . The vector of all non-zero elements in , e.g.…”

“…To compare SEM-BayesC with SEM-GWAS of Momen et al (2018) , we simulated data based on real genotypes from the Rice Diversity Panel. The simulation scenarios in Chen et al (2017) were applied to simulate different genetic architectures.…”

“… Gianola and Sorensen (2004) used SEM to extend conventional multi-trait linear mixed models to accommodate for recursive and simultaneous relationships among traits, which allows traits to be explanatory variables for other traits. Recently, Momen et al (2018 , 2019 ) proposed the SEM-based GWAS (SEM-GWAS) methodology by applying SEM to linear mixed models for GWAS. They showed that while conventional GWAS methodology only provides overall SNP effects, SEM-GWAS can capture the complex causal relationships among traits and further decompose the overall SNP effects into direct and indirect effects.…”

“…The SEM-GWAS method proposed by Momen et al (2018 , 2019 ) is based on linear mixed models with a fixed substitution effect for the tested SNP and a random effect with covariances defined by a “genomic relationship matrix” computed from genotypes ( VanRaden 2008 ) to account for genetic relatedness. Markers are usually implicitly assumed to affect all traits when the ”genomic relationship matrix” is constructed in multi-trait analysis.…”