Translating insights from the seed metabolome into improved prediction for healthful compounds in oat (Avena sativa L.)

Abstract

<jats:title>Abstract</jats:title><jats:p>Oat (<jats:italic>Avena sativa</jats:italic> L.) seed is a rich resource of beneficial lipids, soluble fiber, protein, and antioxidants, and is considered a healthful food for humans. Despite these characteristics, little is known regarding the genetic controllers of variation for these compounds in oat seed. We sought to characterize natural variation in the mature seed metabolome using untargeted metabolomics on 367 diverse lines and leverage this information to improve prediction for seed quality traits. We used a latent factor approach to define unobserved variables that may drive covariance among metabolites. One hundred latent factors were identified, of which 21% were enriched for compounds associated with lipid metabolism. Through a combination of whole-genome regression and association mapping, we show that latent factors that generate covariance for many metabolites tend to have a complex genetic architecture. Nonetheless, we recovered significant associations for 23% of the latent factors. These associations were used to inform a multi-kernel genomic prediction model, which was used to predict seed lipid and protein traits in two independent studies. Predictions for eight of the 12 traits were significantly improved compared to genomic best linear unbiased prediction when this prediction model was informed using associations from lipid-enriched factors. This study provides new insights into variation in the oat seed metabolome and provides genomic resources for breeders to improve selection for health-promoting seed quality traits. More broadly, we outline an approach to distill high-dimensional ‘omics’ data to a set of biologically-meaningful variables and translate inferences on these data into improved breeding decisions.</jats:p>

Improving Genomic Prediction for Seed Quality Traits in Oat (Avena sativa L.) Using Trait-Specific Relationship Matrices

“…In these cases, it may be appropriate to use a combination of dimension-reduction and variable selection methods to select relevant phenotypes or linear combinations of phenotypes. Methods like principal component analysis or factor analysis have been used extensively to cope with high-dimensional traits (Runcie and Mukherjee, 2013 ; Wang and Stephens, 2018 ; Carlson et al, 2019 ; Sakamoto et al, 2019 ; Yu et al, 2019 ; Campbell et al, 2020 ; Rice et al, 2020 ; Runcie et al, 2020 ). These approaches can be used to create derived traits that capture (co)variance in the original data, and marker effects can be easily estimated using GWAS or whole-genome regression approaches.…”

“…Single-nucleotide polymorphism (SNP) data were collected from 11 genotyping experiments for 539 lines (Campbell et al, 2020 ). The approach was used to impute missing marker data (Chan et al, 2016 ).…”

“…The following protocol was used for all experiments that measured fatty acid phenotypes. The methods are described in detail in Campbell et al ( 2020 ) and Carlson et al ( 2019 ). Briefly, dehulled seeds were homogenized, and 100 mg of pulverized tissue was used to separate polar and non-polar compounds using a biphasic extraction method.…”

Selection for seed size has indirectly shaped specialized metabolite abundance in oat (Avena sativaL.)

“…We used two germplasm panels of inbred lines, a diversity panel intended to capture genetic diversity in cultivated oats and an elite panel consisting of lines selected from the North American uniform oat performance nursery. These germplasm panels have been previously described in Campbell et al . (2021) and Hu et al .…”

“…We profiled the seed metabolome in the oat diversity panel and elite panel. Detailed descriptions of extraction and processing of these samples has been previously (Campbell et al ., 2021; Hu et al ., 2021) and is provided here in Method S1 . Briefly, extractions and measurements were conducted at the Bioanalysis and Omics Center of the Analytical Resources Core (“ARC-BIO”), at Colorado State University (Fort Collins, CO, USA).…”