Genotype × Environment Interaction for Nitrate Concentration in Lettuce

High nitrate concentration is a problem when lettuce (Lactuca sativa L.) is grown under low light intensities. Genotypes with low nitrate concentrations have been identified and are being used in breeding programmes. Previous results indicated the occurrence of genotype × environment (GE) interactions. Because of the important influence of light intensity on nitrate accumulation, two types of interactions may be expected: interactions related to daily variation, and those related to annual variation. In the present investigation both types were investigated using eight butterhead genotypes of lettuce which were repeatedly harvested. No daily variation in nitrate concentration and no corresponding GE-interactions were found, irrespective of the level of global radiation. In contrast, a large annual variation and important corresponding GE-interactions were found. Joint regression analysis on environmental means and on physical factors related to light intensity showed a differential response of genotypes to changing environmental conditions. Multiple joint regression on daylength and change in daylength accounted for two-thirds of the interaction variance. However, deviations from regression were still significant indicating non-linearity of the relationship, or, the existence of other environmental factors contributing to GE-interaction.     

AMMI analysis of rainfed lowland rice (Oryza sativa) trials in Nigeria

The additive main effects and multiplicative interaction (AMMI) model is used to analyse the grain yield data of 13 rice genotypes grown in 12 rainfed lowland rice environments. The trials were organized by the International Network for Genetic Evaluation of Rice in Africa (INGER-Africa) and conducted in Nigeria. Main effects due to environments (E), genotypes (G) and G × E interaction were found to be significant (P = 0.001). Cross validation analysis suggested that an AMMI model with one interaction principal component axis (IPCA) was most useful predictively, whereas Gollobs’ test declared two components, IPCA1 and IPCA2, statistically significant (P = 0.01). The IPCAl, however, accounted for most (47.8%) of the G × E sum of squares. Correlation and regression analysis, and relative scatter of genotype and environment points on the AMMI biplot suggest that the interaction partitioned in IPCA1 resulted from differences in the days to flowering among the genotypes. The paper discusses these in relation to the occurrence of Fe toxicity at the test sites and varietal tolerance to the stress.     

Genotype × environment interactions for digestibility traits in silage maize estimated from in vivo measurements with standard sheep

The aim of this work was to investigate the genotype × environment interaction for in vivo digestibility of organic matter and of crude fibre in silage maize evaluated with standard sheep experiments. In order to test the genotype × year interaction, the first experiment consisted of taking data subsets out of a 26-year experiment and evaluating in vivo digestibility traits at Lusignan (France) on numerous maize genotypes. In order to test the genotype × location interaction, the second experiment was a specific one whereby five hybrids were cropped in diverse locations and then evaluated from experiments with sheep, at Lusignan. The variation attributed to genotype × environment (either a year or a location) interaction for in vivo digestibility traits was distinctly lower than the variation due to the main genotypic effect. Therefore, the in vivo digestibility of organic matter and of crude fibre in maize genotypes could be accurately assessed from silages cropped in a simple experimental design, which included replicates, but only a small number of years or locations. This also confirmed the results obtained with in vitro digestibility traits from large multi-environmental designs which highlighted the low importance of genotype × environment interactions and contributed to the validation of in vitro criteria.     

Analysis of embryo, endosperm, cytoplasmic and maternal effects for heterosis of protein and lysine content in indica hybrid rice

The heterosis controlled by genetic main effects and genotype × environment (GE) interaction effects for protein content and lysine content traits of indica hybrid rice, Oryza sativa L., was studied by using a genetic model for quantitative traits of triploid endosperm. The experiment was conducted over 2 years in a factorial design that included nine cytoplasmic male-sterile lines as females and five restorer lines as males. It was revealed that heterosis of protein content and lysine content were simultaneously controlled by genetic main effects and GE interaction effects. Maternal general heterosis and maternal interaction heterosis were observed. Embryo heterosis or cytoplasm heterosis for lysine content and endosperm heterosis for protein content were more important in general heterosis. Embryo interaction heterosis and cytoplasm interaction heterosis were more important for protein content, but endosperm heterosis was only important for lysine content in GE interaction heterosis. It was shown that some indica hybrid crosses had significant positive heterosis for protein content. Negative heterosis for lysine content was observed in most hybrid crosses.     

A mixed model analysis of 10 years of oat evaluation data: use of agronomic information to explain genotype by environment interaction

The versatility of mixed model procedures in investigating large, unbalanced sets of genotype by environment data is illustrated on an historic set of yields from a South Australian oat evaluation program. Information on specific genotypic traits is included in the analysis in order to isolate unexplained genotype by environment interaction.     

Structuring genotype × environment interaction for quality traits in bread wheat, in two multi-location series of trials

Structure of genotype × environment interaction was studied in two series of trials for three quality traits in bread wheat. Two kinds of environments were present in each series of trials: macro-environments defined as locations or location × year combinations and micro-environments induced by diversified cultural practices within each site. For each trait, a simultaneous clustering procedure was used to identify groups of environments which were homogeneous for interaction. An optimised series of trials was proposed from the clusters obtained. The cultural practice based on nitrogen fertilisation seemed to better diversify environments for interaction than use of fungicide, when all quality traits were considered. Determining an optimised series of trials simultaneously for the three traits led to keeping more environments than when one trait was considered. Suggestions for establishing a series of trials for a multi-trait analysis were proposed. This revised version was published online in July 2006 with corrections to the Cover Date.     

Genotype × Environment Interaction in the Andean Grain Crop Quinoa (Chenopodium quinoa) in Temperate Environments

Ten varieties of quinoa with origins ranging from latitude 39° S to 12° S and from sea level to an altitude of 3,800 m were grown on two soil types in two years in Cambridgeshire, England, in order to assess the extent and nature of genotype × environment (G × E) interactions and identify genotypes suited for cultivation at temperate latitudes. There was evidence that varieties differed in their susceptibility to water logging during germination. Plant height was strongly influenced by competition with weeds, and varieties differed in their susceptibility to this. The number of days to anthesis and to maturity were strongly dependent on the variety, but these periods were generally longer following an earlier sowing. The grain yield was also strongly dependent on the variety, but weed competition, a micronutrient deficiency and bird damage affected the varieties differently. Varieties originating at high latitudes gave the highest yields, about 5,000 kg/ha. Earliness and yield were strongly associated at the level of variety means, but the pattern of G × E interaction differed among the variables measured.     

Assessment of genotype × environment interactions for yield and fiber quality in cotton performance trials

Plant breeding programs involving a wide range of crop plants routinely practice selection (directly or indirectly) for genotypes that display stability for a given trait or set of traits across testing environments through the genotype evaluation process. Genotype stability for trait performance is a direct measure of the presence and effect of genotype × environment interactions, which result from the differential performance of a genotype or cultivar across environments. The genotype evaluation process also requires selection of the proper field trial locations that best represent the target environments the breeding program is directed toward. In this study, we assessed the extent to which genotype × environment interactions affected agronomic performance (lint yield, gin turnout) and fiber quality (fiber length, fiber strength, uniformity index, micronaire, fiber elongation) in a series of cotton (Gossypium hirsutum) performance trials in 12 location–year environments in South Carolina. Genotype × environment interactions affecting lint yield were larger in higher yielding environments, while interactions for fiber strength were greater for genotypes with lower mean fiber strength values. Two regions within the South Carolina cotton production areas were identified as proper testing locations for lint yield performance, while testing for fiber strength can be accomplished in any location within the statewide cotton production areas. The U.S. Government's right to retain a non-exclusive, royalty-free license in and to any copyright is acknowledged.     

Genomic selection allowing for marker‐by‐environment interaction

Genomic selection has been routinely implemented in plant breeding in two stages. The first stage usually omits the marker information and estimates adjusted means of genotypes across environments. The second stage uses the adjusted means to predict genomic breeding values. However, if the effects of markers vary substantially between different environments, it may be important to account for this variation for varieties adapted to different environments. Using two maize data sets, we investigated whether modelling the marker‐by‐environment interaction can improve the predictive ability of genomic selection relative to modelling genotype‐by‐environment interaction alone. Modelling the marker‐by‐environment interaction did not substantially increase the predictive ability relative to modelling only the genotype‐by‐environment interaction for the two tested data sets. Thus, genomic selection, carried out in a stagewise fashion, such that the marker information is omitted until the last stage of the process, may suffice for most practical purposes. Moreover, predictive ability did not reduce substantially even when the number of markers with consistent effects across environments used for genomic prediction was reduced to about 50.     

Prediction of barley progeny performance in the presence of genotype environment interaction

Twenty recombinant inbred line (RIL) populations of European two‐row spring barley and their parents were tested in six environments in the Netherlands to investigate the prediction of progeny yield level, yield variance, stability level and stability variance, based on parent information. Progeny yield level is positively correlated with midparent value for average yield. Progeny yield variance is more difficult to predict, but there does appear to be a promising negative correlation between progeny yield variance and Habgood's (1977) parental similarity measure. To quantify yield stability, three statistics were calculated: Finlay and Wilkinson's (1963) regression coefficient b_i, Shukla's (1972) stability variance σs_i² and Eberhart and Russell's (1966) mean squared deviation d_i². The first stability statistic describes a different aspect of the response pattern to change in environment from the last two. Parents with high b_i values appear to have a better average yield, i.e. they react more positively to an improvement in the environment than the other genotypes. The average b_i value of the progeny is positively correlated with the midparent value, indicating its heritable nature. There are also indications that d_i² and σ_i² are heritable but their repeatability is poor. Therefore, it is concluded that only prediction of bi is useful in practical plant breeding. There is a positive correlation between progeny yield variance and progeny variance for b_i but we conclude that the inaccuracy of the stability variance estimates is too high for good predictors for progeny stability variance to be found.     

Pattern analysis of genotype × environment interaction for striga resistance and grain yield in African sorghum trials

The parasitic weed Striga hermonthica (Del.) Benth. seriously limits sorghum [Sorghum bicolor (L.) Moench] production in Sub-Saharan Africa. As an outbreeder, S. hermonthica is highly variable with an extraordinary capacity to adapt to different hosts and environments, thereby complicating resistance breeding. To study genotype x environment (G x E) interaction for striga resistance and grain yield, nine sorghum lines, 36 F₂ populations and five local checks were grown under striga infestation at two locations in both Mali and Kenya. Mean squares due to genotypes and G x E interaction were highly significant for both sorghum grain yield and area under striga severity progress curve(ASVPC, a measure of striga emergence and vigor throughout the season). For grain yield, the entry x location-within-country interaction explained most of the total G x E while for ASVPC, entry x country and entry x location-within-country interactions were equally important. Pattern analysis (classification and ordination techniques) was applied to the environment-standardized matrix of entry x environment means. The classification clearly distinguished Malian from Kenyan locations for ASVPC, but not for grain yield. Performance plots for different entry groups showed differing patterns of adaptation. The ordination biplot underlined the importance of entry x country interaction for ASVPC. The F₂ derived from the cross of the striga-resistant line Framida with the striga-tolerant cultivar Seredo was the superior entry for both grain yield and ASVPC, underlining the importance of combining resistance with tolerance in striga resistance breeding. The observed entry x country interaction for ASVPC may be due to the entries' different reactions to climatic conditions and putative differences in striga virulence in Mali and Kenya. This revised version was published online in August 2006 with corrections to the Cover Date.     

Analysing genotype by environment interaction in Dutch potato variety trials using factorial regression

Summary Genotype by environment interaction was investigated for yield data from the official Dutch Variety List trials for potato. The data set included 64 genotypes by 26 environments, where environments consisted of year by soil type combinations. Factorial regression models incorporating genotypic and environmental covariates in the interaction were used to analyse the data. The merits of factorial regression models were compared with those of biadditive models. Factorial regression models and biadditive models described comparable amounts of interaction, but factorial regression models provided a better basis for biological interpreration of the interaction.This article was previously published in Euphytica 82: 149–155.     

Analysing genotype by environment interaction in Dutch potato variety trials using factorial regression

Summary Genotype by environment interaction was investigated for yield data from the official Dutch Variety List trials for potato. The data set included 64 genotypes by 26 environments, where environments consisted of year by soil type combinations. Factorial regression models incorporating genotypic and environmental covariates in the interaction were used to analyse the data. The merits of factorial regression models were compared with those of biadditive models. Factorial regression models and biadditive models described comparable amounts of interaction, but factorial regression models provided a better basis for biological interpretation of the interaction.     

水稻稻米Hg累积的遗传生态特性研究

选用6个不同类型的水稻品种（系）,按完全双列杂交设计（6×5）配成一套亲本、F1和F2 3个世代的遗传材料。采用包括种子、细胞质、母体植株三套遗传体系的种子性状遗传模型和统计分析方法,系统分析了稻米汞含量性状的遗传特点。主要结果如下：遗传方差分析结果表明,Hg元素含量除了受制于种子基因效应、细胞质基因效应和母体植株基因效应等遗传主效应外,还会明显受到各遗传效应与环境互作效应的影响,汞元素含量主要以遗传主效应为主。在各遗传体系中,汞元素含量以显性效应及其与环境互作效应为主,种子基因效应也有一定作用。遗传率和选择响应分析结果表明,汞总狭义遗传率较高,世代的遗传传递力较强,在鉴定筛选淘汰时,需要加大群体才能达到降低Hg元素含量的育种目标。     

Studying the effect of environmental variables on the genotype × environment interaction of tomato

Genotype × environment interaction (GEI) affects marketable fruit yield and average fruit weight of both hybrid and open-pollinated (OP) tomato genotypes. Cultivars vary significantly for marketable fruit yield, with hybrid cultivars having, on average, higher yield than OP cultivars. However, information is scanty on environmental factors affecting the differential response of tomato genotypes across environments. Hence, the aim of this research was to use factorial regression (FR) and partial least squares (PLS) regression, which incorporate external environmental and genotypic covariables directly into the model for interpreting GEI. In this research, data from an FAO multi-environment trial comprising 15 tomato genotypes (7 hybrid and 8 OP) evaluated in 18 locations of Latin America and the Caribbean were analyzed using FR and PLS. Environmental factors such as days to harvest, soil pH, mean temperature (MET), potassium available in the soil, and phosphorus fertilizer accounted for a sizeable portion of GEI for marketable fruit yield, whereas trimming, irrigation, soil organic matter, and nitrogen and phosphorus fertilizers were important environmental covariables for explaining GEI of average fruit weight. Locations with relatively high minimum and mean temperatures favored the marketable fruit yield of OP heat-tolerant lines CL 5915-223 and CL 5915-93. An OP cultivar (Catalina) and a hybrid (Apla) showed average marketable fruit yield across environments, while two hybrids (Sunny and Luxor) exhibited outstanding marketable fruit yield in high yielding locations (due to lower temperatures and higher pH) but a sharp yield loss in poor environments. Two stable hybrid genotypes in high yielding environments, Narita and BHN-39, also showed high and stable yield in average and low yielding environments.     

Mixed models including environmental covariables for studying QTL by environment interaction

The study of the phenotypic responses of a set of genotypes in their dependence on the environment has always been an important area of research in plant breeding. Non-parallelism of those responses is called genotype by environment interaction (GEI). GEI especially affects plant breeding strategies, when the phenotypic superiority of genotypes changes in relation to the environment. The study of the genetic basis of GEI involves the modelling of quantitative trait locus (QTL) expression in its dependence on environmental factors. We present a modelling framework for studying the interaction between QTL and environment, using regression models in a mixed model context. We integrate regression models for QTL main effect expression with factorial regression models for genotype by environment interaction, and, in addition, take care to model adequately the residual genetic variation. Factorial regression models describe GEI as differential genotypic sensitivity to one or more environmental covariables. We show how factorial regression models can be generalized to make also QTL expression dependent on environmental covariables. As an illustrative example, we reanalyzed yield data from the North American Barley Genome Project. QTL by environment interaction for yield, as identified at the 2H chromosome could be described as QTL expression in relation to the magnitude of the temperature range during heading. This revised version was published online in July 2006 with corrections to the Cover Date.     

Repeatability of different stability parameters for grain yield in chickpea

The presence of genotype × environment (GE) interactions in plant breeding experiments has led to the development of several stability parameters in the past few decades. The present study investigated the repeatability of these parameters for 16 chickpea (Cicer arietinum L.) genotypes by correlating their estimates obtained from extreme subsets of environments within a year and also over years. Based on the estimates of response and stability parameters within each trial, the ranking of genotypes in the low-yielding subset differed from that in the high-yielding subset. This indicates poor repeatability for response and stability parameters over the extreme environmental subsets. The estimates of mean yield and stability parameters represented by ecovalence, W²_i, were consistent over years, whereas those of response parameters (b_i, and S²_i) showed poor repeatability. Our results suggest that single-year results for yield and stability can be used effectively for selecting cultivars with stable grain yield if tested in a wider range of environments.     

Contributions to the Analysis of Genotype × Environment Interactions: Experimental Results of the Application and Comparison of Clustering Techniques for the Stratification of Field Test Sites

The results from multienvironment field performance trials of cultivars are usually analysed as two‐way classification data with rows=genotypes/cultivars and columns=environments (locations and/or years). To reduce the impact of genotype × environment interaction effects, one commonly stratifies genotypes/cultivars or environments by cluster analysis techniques into homogeneous groups so that interactions within groups are minimized. By such a stratification, for example of test sites, with regard to similarity of genotype × environment interactions and the selection of only one representative test site from each group, the overall number of necessary test sites for yield trials can be reduced. In the literature, many clustering techniques have been proposed. Systematic comparisons between different cluster methods, however, are rather rare. A single cluster method is characterized by `measure of distance', `stopping criterion', `algorithm' and `level of significance'. In this paper, 11 clustering techniques were applied to extensive yield data sets of several agricultural crops (faba bean, fodder beet, oat, winter oilseed rape and sugar beet) from the official registration trials of the German `Bundessortenamt'. The results were compared with each other using two proposed parameters: measure of resemblance (for two classifications) and cluster size (for one classification). Neither the level of significance nor the algorithm has a substantial impact on the resulting clusters. The final results of clustering are therefore mainly determined by the stopping criterion with its associated measure of distance. If one uses tests for crossover interactions as stopping criteria, the resulting clusters are larger than the resulting clusters for the F‐test of conventional interactions in an analysis of variance. The cluster size decreases with increasing sensitivity of the tests that are used as stopping criteria. Finally, recommendations for the choice and handling of clustering techniques for practical applications are given.     

Contributions to the Analysis of Genotype × Environment Interactions: Theoretical Results of the Application and Comparison of Clustering Techniques for the Stratification of Field Test Sites

For an analysis of cross-classified data sets with rows = genotypes and columns = environments (locations and/or years), existing genotype × environment interactions are of major importance. Differential responses of genotypes (environments) across environments (genotypes) are expressed by these effects. To reduce the impact of these genotype × environment interaction effects, one commonly stratifies genotypes or environments by cluster analysis techniques into homogeneous groups so that interactions within groups are minimized. The present paper presents a comprehensive overview of the numerous procedures for stratification of genotypes or environments by cluster analysis which have been proposed in the literature. In these studies, two different concepts of interaction have been used: the crossover concept of interaction [different rank orders of genotypes (environments) within environments (genotypes)] or the usual statistical concept of interaction (deviations from additivity of main effects in the linear model). For a quantitative characterization of cluster techniques and for a comparison of two different clustering procedures, two parameters are introduced and discussed: measure of resemblance for two classifications and cluster size for one classification.     

Prediction of indirect selection for seed and forage yield of lucerne based on evaluation under spaced planting

Phenotypic selection under spaced planting is frequently used in forage species. This study aimed to compare the predicted efficiency of direct selection for lucerne seed or forage yield under dense planting with indirect selection based on the same traits or seed yield components evaluated under spaced planting. Sixteen genotypes randomly chosen from a representative sample of locally adapted germplasm were grown for two years as individual clones spaced at 75 cm (density =1.78 plants/m²) and in dense plots formed by one row of four clones spaced at 10 cm (density = 50 plants/m²) using a randomized complete block design with three replications. Indirect selection based on seed yield under spaced planting was just 19% less efficient than direct selection, owing to moderate genetic correlation between plant densities (r_g = 0.66) and somewhat higher broad‐sense heritability under spaced planting than under dense planting. The relative efficiency of indirect selection for seed yield in density based on individual seed yield components under spaced planting ranged from modest to very low and was always below 45%. The efficiency of indirect selection for dry matter yield based on yield response under spaced planting was moderate for total yield (64%) and very low for second‐year yield (32%) relative to direct selection.