A multi‐trait multi‐locus stepwise approach for conducting GWAS on correlated traits

Abstract

The ability to accurately quantify the simultaneous effect of multiple genomic loci on multiple traits is now possible due to current and emerging high-throughput genotyping and phenotyping technologies. To date, most efforts to quantify these genotypeto-phenotype relationships have focused on either multi-trait models that test a single marker at a time or multi-locus models that quantify associations with a single trait. Therefore, the purpose of this study was to compare the performance of a multi-trait, multi-locus stepwise (MSTEP) model selection procedure we developed to (a) a commonly used multi-trait single-locus model and (b) a univariate multilocus model. We used real marker data in maize (Zea mays L.) and soybean (Glycine max L.) to simulate multiple traits controlled by various combinations of pleiotropic and nonpleiotropic quantitative trait nucleotides (QTNs). In general, we found that both multi-trait models outperformed the univariate multi-locus model, especially when analyzing a trait of low heritability. For traits controlled by either a combination of pleiotropic and nonpleiotropic QTNs or a large number of QTNs (i.e., 50), our MSTEP model often outperformed at least one of the two alternative models.When applied to the analysis of two tocochromanol-related traits in maize grain, MSTEP identified the same peak-associated marker that has been reported in a previous study. We therefore conclude that MSTEP is a useful addition to the suite of statistical models that are commonly used to gain insight into the genetic architecture of agronomically important traits.

Genome-Wide Association Insights into the Genomic Regions Controlling Oil Production Traits inAcrocomia aculeata(neotropical native palm)

“…The multi-trait analysis was fitted using a multivariate stepwise method (MSTEP) implemented in the software TASSEL, being conducted in all pairwise trait combinations (Fernandes et al, 2022). Prior to running MSTEP, each trait was normal quantile transformed using the orderNorm function from the R package bestNormalize (Peterson, 2020).…”

“…Many software packages have implemented multi-trait GWAS models (Furlotte & Eskin, 2015; Pritikin et al, 2021; Zhou & Stephens, 2014; Bradbury et al, 2007). Especially when analyzing low heritability traits, multi-trait models can outperform univariate multi-locus models (Fernandes et al, 2022).…”

Genomic prediction of cereal crop architecture traits using models informed by gene regulatory circuitries from maize

“…We implemented two multi-locus stepwise model selection procedures to identify markers exhibiting strong statistical associations with TBN and LA in maize. The first procedure was the multi-trait, multilocus (MSTEP) procedure, which is described in detail in Fernandes et al (2022). Briefly, this procedure fits a series of multi-trait, multi-locus models in a stepwise manner to identify markers exhibiting strong additive associations with multiple traits.…”

Pleiotropic genetic association analysis with multiple phenotypes using multivariate response best-subset selection

“…For example, some distance regression model methods have been proposed to perform association test based on the similarity matrices of genotype and phenotype [ 21 – 24 ]. Samuel et al [ 25 ] developed a multi-trait, multi-locus stepwise model selection procedure that proves to be valuable in scenarios where phenotypes are influenced by both pleiotropic and non-pleiotropic quantitative trait nucleotides. Bottolo et al [ 26 ] proposed a Bayesian variable selection to perform polygenic association with multiple phenotypes, and exploited parallel processing framework for fully multivariate modelling of groups of correlated phenotypes at the genome-wide scale.…”