Economical optimization of a breeding scheme by selective phenotyping of the calibration set in a multi-trait context: application to bread making quality.

KEY MESSAGE: Trait-assisted genomic prediction approach is a way to improve genetic gain by cost unit, by reducing budget allocated to phenotyping or by increasing the program's size for the same budget. This study compares different strategies of genomic prediction to optimize resource allocation in breeding schemes by using information from cheaper correlated traits to predict a more expensive trait of interest. We used bread wheat baking score (BMS) calculated for French registration as a case study. To conduct this project, 398 lines from a public breeding program were genotyped and phenotyped for BMS and correlated traits in 11 locations in France between 2000 and 2016. Single-trait (ST), multi-trait (MT) and trait-assisted (TA) strategies were compared in terms of predictive ability and cost. In MT and TA strategies, information from dough strength (W), a cheaper trait correlated with BMS (r = 0.45), was evaluated in the training population or in both the training and the validation sets, respectively. TA models allowed to reduce the budget allocated to phenotyping by up to 65% while maintaining the predictive ability of BMS. TA models also improved the predictive ability of BMS compared to ST models for a fixed budget (maximum gain: + 0.14 in cross-validation and + 0.21 in forward prediction). We also demonstrated that the budget can be further reduced by approximately one fourth while maintaining the same predictive ability by reducing the number of phenotypic records to estimate BMS adjusted means. In addition, we showed that the choice of the lines to be phenotyped can be optimized to minimize cost or maximize predictive ability. To do so, we extended the mean of the generalized coefficient of determination (CDmean) criterion to the multi-trait context (CDmulti).

Optimization of multi-environment trials for genomic selection based on crop models.

KEY MESSAGE: We propose a statistical criterion to optimize multi-environment trials to predict genotype × environment interactions more efficiently, by combining crop growth models and genomic selection models. Genotype × environment interactions (GEI) are common in plant multi-environment trials (METs). In this context, models developed for genomic selection (GS) that refers to the use of genome-wide information for predicting breeding values of selection candidates need to be adapted. One promising way to increase prediction accuracy in various environments is to combine ecophysiological and genetic modelling thanks to crop growth models (CGM) incorporating genetic parameters. The efficiency of this approach relies on the quality of the parameter estimates, which depends on the environments composing this MET used for calibration. The objective of this study was to determine a method to optimize the set of environments composing the MET for estimating genetic parameters in this context. A criterion called OptiMET was defined to this aim, and was evaluated on simulated and real data, with the example of wheat phenology. The MET defined with OptiMET allowed estimating the genetic parameters with lower error, leading to higher QTL detection power and higher prediction accuracies. MET defined with OptiMET was on average more efficient than random MET composed of twice as many environments, in terms of quality of the parameter estimates. OptiMET is thus a valuable tool to determine optimal experimental conditions to best exploit MET and the phenotyping tools that are currently developed.

Grain characterization and milling behaviour of near-isogenic lines differing by hardness.

Wheat grain hardness is a major factor affecting the milling behaviour and end-product quality although its exact structural and biochemical basis is still not understood. This study describes the development of new near-isogenic lines selected on hardness. Hard and soft sister lines were characterised by near infrared reflectance (NIR) and particle size index (PSI) hardness index, grain protein content, thousand kernel weight and vitreousness. The milling behaviour of these wheat lines was evaluated on an instrumented micromill which also measures the grinding energy and flour particle size distribution was investigated by laser diffraction. Endosperm mechanical properties were measured using compression tests. Results pointed out the respective effect of hardness and vitreousness on those characteristics. Hardness was shown to influence both the mode of fracture and the mechanical properties of the whole grain and endosperm. Thus, this parameter also acts on milling behaviour. On the other hand, vitreousness was found to mainly play a role on the energy required to break the grain. This study allows us to distinguish between consequences of hardness and vitreousness. Hardness is suggested to influence the adhesion forces between starch granules and protein matrix whereas vitreousness would rather be related to the endosperm microstructure.

Effects of food deprivation and particle size of ground wheat on digestibility of food components in broilers fed on a pelleted diet.

The first aim of the experiment was to study the effect of wheat (Triticum aestivum) particle size on the digestibility of starch in a pelleted diet given to broilers. The second aim was to study the consequences of food deprivation before the excreta collection period (from 21 to 24 d). Wheat from a strong hardness cultivar was incorporated at 546.1 g/kg in diets. The other main ingredients were soybean meal (353.5 g/kg) and rapeseed oil (55.0 g/kg). Diets were given as pellets. The experimental design was a 2 x 2 factorial design testing two particle sizes of wheat flour and two procedures of a balance experiment (with or without food deprivation). Birds given diet C (wheat coarse grinding before pelleting) had significantly greater gizzard weight than birds fed on diet F (wheat fine grinding before pelleting). Starch digestibility value was significantly increased when birds were fed on diet F. This effect was halved by food deprivation. No significant effect of grain particle size was observed for protein and lipid digestibility values. However, food deprivation decreased apparent protein digestibility, with an effect which was more pronounced for fine than for coarse grinding. AMEN of the diet was significantly improved by fine grinding of wheat and decreased by food deprivation. However, no significant differences in growth performance were induced by differences in wheat grinding. No significant effect of grinding was observed on the water excretion:feed intake ratio. No significant difference was observed for vent score between treatments. There was over-excretion of starch in the first hours of refeeding following food deprivation.

Relationships between digestibilities of food components and characteristics of wheats (Triticum aestivum) introduced as the only cereal source in a broiler chicken diet.

1. The aim of the experiment was to establish relationships between chemical or physical characteristics of wheats (Triticum aestivum) and digestibilities of food components in broiler chickens fed on wheat-based diets. Twenty-two wheat samples, each differing by their cultivar origin, were included at 550 g/kg in diets offered to male Ross broiler chicks. The other main ingredients were soya bean meal (340 g/kg) and rapeseed oil (68.5 g/kg). Diets were given as pellets. 2. In vitro viscosities of wheats measured as potential applied viscosity (PAV) or real applied viscosity (RAV) varied between 1.91 and 6.03, or between 0.95 and 3.81 ml/g (dry matter basis), respectively. Hardness of wheats varied between 17 (soft) and 95 (very hard), and lipase activity of wheats varied from 1 to 13.6 (relative scale). 3. PAV and RAV values were not significantly correlated with hardness. PAV and RAV values were correlated with (80:20) ethanol:water insoluble, water soluble arabinoxylans (r = 0.961, 0.932, respectively), with the amount of water retained by cell walls (r = 0.656, 0.492, respectively), and with lipase activity (r = 0.600, 0.532, respectively. 4. Hardness was correlated with ash (r = -0.484), nitrogen (r = 0.534), mean particle size of wheat flours (r = 0.631), and specific energy of pelleting (r = -0.574). 5. Wheat diets were evaluated in two assays with 3-week-old chicks, with 11 diets per assay. In each assay, a balance experiment was carried out using the total collection method. Growth performance was also measured during the balance experiment. 6. In vitro viscosity parameters were negatively correlated with diet AMEn (P < 0.05), lipid digestibility (P < 0.05) and, to a lesser extent, protein digestibility (P < 0.05). In vitro viscosity data were positively correlated with food:gain ratio (P < 0.05) and water loss parameters (P < 0.05), and were not significantly (P > 0.05) correlated with starch digestibility. 7. Wheat hardness-related parameters were correlated (P < 0.01) with individual starch digestibility (hardness, proportion of coarse particles in wheat flour, specific energy of pelleting: r = -0.273, -0.305, 0.212, respectively). 8. Wheat lipase activity was negatively correlated with individual lipid (r = -0.179; P < 0.05) and starch (r = -0.225; P < 0.01) digestibilities and with individual diet AMEn (r = -0.266; P < 0.001). Individual diet AMEn values were correlated (r = 0.175) with the values calculated by the EU AMEn prediction equation (Fisher and McNab, 1987). Among the correlations observed between the individual measured AMEn:EU predicted AMEn ratio and wheat parameters (P < 0.05), the correlation obtained with wheat lipase was the highest (r = -0.195). The correlations with lipase could be explained in part by strong correlations between lipase and in vitro viscosity parameters.