Genomic Prediction and Selection for Fruit Traits in Winter Squash

Abstract

Improving fruit quality is an important but challenging breeding goal in winter squash. Squash breeding in general is resource-intensive, especially in terms of space, and the biology of squash makes it difficult to practice selection on both parents. These restrictions translate to smaller breeding populations and limited use of greenhouse generations, which in turn, limits genetic gain per breeding cycle and increases cycle length. Genomic selection is a promising technology for improving breeding efficiency; yet, few studies have explored its use in horticultural crops. We present results demonstrating the predictive ability of whole-genome models for fruit quality traits. Predictive abilities for quality traits were low to moderate, but sufficient for implementation. To test the use of genomic selection for improving fruit quality, we conducted three rounds of genomic recurrent selection in a butternut squash (Cucurbita moschata) population. Selections were based on a fruit quality index derived from a multi-trait genomic selection model. Remnant seed from selected populations was used to assess realized gain from selection. Analysis revealed significant improvement in fruit quality index value and changes in correlated traits. This study is one of the first empirical studies to evaluate gain from a multi-trait genomic selection model in a resource-limited horticultural crop.

Gain from genomic selection for a selection index in two‐row spring barley

“…These results likely are due to the large index weight and substantial genetic gain for PHS. Gain in SI value and some component traits after calculating SI with GEBVs has been previously reported (Hernandez et al, 2020;Tiede & Smith, 2018).…”

“…Empirical experiments that evaluated realized genetic gain from GS in plants suggest GS and PS may perform similarly (Rutkoski et al., 2015; Sallam & Smith, 2016) but reduced cycle time in GS leads to greater gain per unit time. Use of a SI in GS may result in gain for component traits (Combs & Bernardo, 2013), index value (Massman et al., 2013), or both for positively (Hernandez et al., 2020) and negatively genetic correlated traits (Tiede & Smith, 2018). Prediction accuracy for SI value may be higher with GS than PS even if single trait accuracies are similar between the two methods (Heffner et al., 2011).…”

Genetic characterization of carrot root shape and size using genome-wide association analysis and genomic-estimated breeding values

“…In addition to GWAS, marker data were used to calculate GEBVs, using best linear unbiased predictors (BLUPs) (Henderson 1963). First, the marker matrix used for estimation of kinship was thinned significantly: markers were thinned to a maximum density of one marker per 1 kb, with no missing data, resulting in 12,370 SNPs, with an average distance between SNPs of 36.55 kb; this marker density is sufficient to support maximal predictive ability, and thus was an appropriate base marker set for all subsequent analyses, as it allowed for each prediction accuracy-limiting variable to be evaluated individually.…”

Ideas in Genomic Selection with the Potential to Transform Plant Molecular Breeding: A Review

“…Across all of these papers there was consensus that while the best specific GS modeling approach varied depending on the trait architecture, GS models in general were capable of predicting many agronomically important traits with a high enough accuracy compared to phenotypic selection to warrant incorporation into plant breeding programs. Since this review in 2017, the number of studies demonstrating the predictive capabilities of GS has increased and expanded from previously studied crops such as wheat (Arruda et al 2016;Belamkar et al 2018;Haile et al 2018;Lozada et al 2019), pea (Annicchiarico et al 2019, and alfalfa (Biazzi et al 2017), to a wider variety of plant species including sunflower (Dimitrijevic and Horn 2018), coffee (Sousa et al 2019), cassava (Wolfe et al 2017), strawberry (Gezan et al 2017), winter squash (Hernandez et al 2020), and cranberry (Covarrubias-Pazaran et al 2018). Cost-benefit analyses of incorporating GS into breeding programs were also discussed and indicated that GS can be substantially cheaper than phenotypic selection and brings the advantage that genotypes can be used for predicting multiple traits.…”