Explore the vast range of titles available.

KEY MESSAGE : New methods that incorporate the main and interaction effects of high-dimensional markers and of high-dimensional environmental covariates gave increased prediction accuracy of grain yield in wheat across and within environments. In most agricultural crops the effects of genes on traits are modulated by environmental conditions, leading to genetic by environmental interaction (G × E). Modern genotyping technologies allow characterizing genomes in great detail and modern information systems can generate large volumes of environmental data. In principle, G × E can be accounted for using interactions between markers and environmental covariates (ECs). However, when genotypic and environmental information is high dimensional, modeling all possible interactions explicitly becomes infeasible. In this article we show how to model interactions between high-dimensional sets of markers and ECs using covariance functions. The model presented here consists of (random) reaction norm where the genetic and environmental gradients are described as linear functions of markers and of ECs, respectively. We assessed the proposed method using data from Arvalis, consisting of 139 wheat lines genotyped with 2,395 SNPs and evaluated for grain yield over 8 years and various locations within northern France. A total of 68 ECs, defined based on five phases of the phenology of the crop, were used in the analysis. Interaction terms accounted for a sizable proportion (16 %) of the within-environment yield variance, and the prediction accuracy of models including interaction terms was substantially higher (17–34 %) than that of models based on main effects only. Breeding for target environmental conditions has become a central priority of most breeding programs. Methods, like the one presented here, that can capitalize upon the wealth of genomic and environmental information available, will become increasingly important.

L'analyse de données joue un rôle croissant dans la recherche agronomique, dans l'expertise scientifique et dans les études prospectives. Cet ouvrage, conçu comme un guide méthodologique, présente les intérêts et les limites de différentes méthodes statistiques permettant d'analyser des données agronomiques issues de réseaux expérimentaux et de réaliser des méta-analyses.

A large network of field experiments has been conducted over several years across France to identify combinations of winter wheat cultivars and management practices in which partial resistances under limited chemical protection would achieve adequate disease management, while leading to satisfactory yield performance, and so achieve the double objective of ecological sustainability and economic viability. Little information is available to document the variation in multiple disease levels, a n ecessary step towards a chemical extensification process, in wheat networked experiments. This article provides a description of dis- ease intensities in a set of 101 experiments totalling 3525 individual wheat plots over eight successive years (2003 – 2010). The diseases considered are brown rust (BR, Puccinia triticina ), yellow rust (YR, Puccinia striiformis ), fusarium head blight (FHB, Fusarium graminearum , F. culmorum ,and F. avenaceum ), pow- dery mildew (PM, Blumeria graminis ), and septoria tritici blotch (STB, Zymoseptoria tritici ). Hierarchical cluster analysis led to the identification of three variety groups associated with (1) moderate-low disease levels in general, except for YR (moderate levels) – 16 varie- ties; (2) moderate-low BR, YR, and FHB levels, and moderate PM and STB levels – 12 varieties; (3) com- paratively higher BR, YR, FHB, and STB levels, and moderate PM levels – 17 varieties. The association of disease levels represented as binary categories (i.e., epidemics vs. non-epidemics) with climatic years corresponded to chi-square values ( χ 2 =87.0 – 1402) that were one to two orders of magnitude larger than the values corresponding to the associations of diseases with variety groups ( χ 2 =6.41 – 321) or with levels of crop management ( χ 2 =21.2 – 82.1). Multivariate non parametric analyses indicated the existence of three disease syndromes, two of which being dominated by BR or STB, and a third associated with diverse diseases and frequent FHB. This suggests that STB and BR might each be considered as key-stone species dominat- ing specific wheat disease syndromes. Multiple corre- spondence analysis highlighted the linkages between multiple epidemic occurrence and the three character- ized variety groups. Risk factors analyses conducted through logistic regressions provided quantitative estimates of the contribution of climatic years, variety groups, and crop management, to the likelihood of epidemic occurrence for each of the five diseases considered. The results indicate that climatic years, wheat varieties, and crop management, in this decreasing order, define disease epidemic risk in the multiple wheat-diseases pathosystem.

A data set generated by a multi-year (2003 – 2010) and multi-site network of experiments on winter wheat varieties grown at different levels of crop manage- ment is analysed in order to assess the importance of climate on the variability of wheat health. Wheat health is represented by the multiple pathosystem involving five components: leaf rust, yellow rust, fusarium head blight, powdery mildew, and septoria tritici blotch. An overall framework of associations between multiple diseases and climate variables is developed. This framework involves disease levels in a binary form (i.e. epidemic vs. non- epidemic) and synthesis variables accounting for climate over spring and early summer. The multiple disease- climate pattern of associations of this framework con- forms to disease-specific kn owledge of climate effects on the components of the pathosystem. It also concurs with a (climate-based) risk factor approach to wheat diseases. This report emphasizes the value of large scale data in crop health assessment and the usefulness of a risk factor approach for both tactical and s trategic decisions for crop health management.

Septoria leaf blotch is a widespread disease caused by the fungus Zymoseptoria tritici (formely known as Mycosphaerella graminicola). It causes yield losses in winter wheat crops (Triticum aestivum L.) in many European countries. In this study, we aimed to develop statistical models for estimating regional and site-specific incidence of Septoria leaf blotch from in-season field measurements. Four generalised linear models and four generalised linear mixed-effect models were fitted to six years of data collected from a major wheat-producing area of France, using frequentist and Bayesian methods. We compared the abilities of these models to predict S. tritici incidence over different time scales. We found that the best models were those that included site-year effects and disease risk ratings based on sowing dates and cultivar resistance levels. These models can be used to estimate the dynamics of disease incidence from observations collected in regional surveys and, as such, could help regional extension services evaluate current disease incidence at the regional scale. The proposed models could also be adjusted to make use of site-specific in-season field measurements for the estimation of site-specific disease incidence. With the current survey design, site-specific estimates are more accurate than regional estimates after mid-May. Such estimates could be used to help farmers adapt their control strategies locally during the growing season.

A large network of field experiments has been conducted over several years across France to identify combinations of winter wheat cultivars and management practices in which partial resistances under limited chemical protection would achieve adequate disease management, while leading to satisfactory yield performance, and so achieve the double objective of ecological sustainability and economic viability. Little information is available to document the variation in multiple disease levels, a n ecessary step towards a chemical extensification process, in wheat networked experiments. This article provides a description of dis- ease intensities in a set of 101 experiments totalling 3525 individual wheat plots over eight successive years (2003 – 2010). The diseases considered are brown rust (BR, Puccinia triticina ), yellow rust (YR, Puccinia striiformis ), fusarium head blight (FHB, Fusarium graminearum , F. culmorum ,and F. avenaceum ), pow- dery mildew (PM, Blumeria graminis ), and septoria tritici blotch (STB, Zymoseptoria tritici ). Hierarchical cluster analysis led to the identification of three variety groups associated with (1) moderate-low disease levels in general, except for YR (moderate levels) – 16 varie- ties; (2) moderate-low BR, YR, and FHB levels, and moderate PM and STB levels – 12 varieties; (3) com- paratively higher BR, YR, FHB, and STB levels, and moderate PM levels – 17 varieties. The association of disease levels represented as binary categories (i.e., epidemics vs. non-epidemics) with climatic years corresponded to chi-square values ( χ 2 =87.0 – 1402) that were one to two orders of magnitude larger than the values corresponding to the associations of diseases with variety groups ( χ 2 =6.41 – 321) or with levels of crop management ( χ 2 =21.2 – 82.1). Multivariate non parametric analyses indicated the existence of three disease syndromes, two of which being dominated by BR or STB, and a third associated with diverse diseases and frequent FHB. This suggests that STB and BR might each be considered as key-stone species dominat- ing specific wheat disease syndromes. Multiple corre- spondence analysis highlighted the linkages between multiple epidemic occurrence and the three character- ized variety groups. Risk factors analyses conducted through logistic regressions provided quantitative estimates of the contribution of climatic years, variety groups, and crop management, to the likelihood of epidemic occurrence for each of the five diseases considered. The results indicate that climatic years, wheat varieties, and crop management, in this decreasing order, define disease epidemic risk in the multiple wheat-diseases pathosystem.