Prediction of maize single-cross hybrid performance: support vector machine regression versus best linear prediction

Accurate prediction of the phenotypic performance of a hybrid plant based on the molecular fingerprints of its parents should lead to a more cost-effective breeding programme as it allows to reduce the number of expensive field evaluations. The construction of a reliable prediction model requires a representative sample of hybrids for which both molecular and phenotypic information are accessible. This phenotypic information is usually readily available as typical breeding programmes test numerous new hybrids in multi-location field trials on a yearly basis. Earlier studies indicated that a linear mixed model analysis of this typically unbalanced phenotypic data allows to construct ɛ-insensitive support vector machine regression and best linear prediction models for predicting the performance of single-cross maize hybrids. We compare these prediction methods using different subsets of the phenotypic and marker data of a commercial maize breeding programme and evaluate the resulting prediction accuracies by means of a specifically designed field experiment. This balanced field trial allows to assess the reliability of the cross-validation prediction accuracies reported here and in earlier studies. The limits of the predictive capabilities of both prediction methods are further examined by reducing the number of training hybrids and the size of the molecular fingerprints. The results indicate a considerable discrepancy between prediction accuracies obtained by cross-validation procedures and those obtained by correlating the predictions with the results of a validation field trial. The prediction accuracy of best linear prediction was less sensitive to a reduction of the number of training examples compared with that of support vector machine regression. The latter was, however, better at predicting hybrid performance when the size of the molecular fingerprints was reduced, especially if the initial set of markers had a low information content.     

Support vector machine regression for the prediction of maize hybrid performance

Accurate prediction of the phenotypical performance of untested single-cross hybrids allows for a faster genetic progress of the breeding pool at a reduced cost. We propose a prediction method based on ɛ-insensitive support vector machine regression (ɛ-SVR). A brief overview of the theoretical background of this fairly new technique and the use of specific kernel functions based on commonly applied genetic similarity measures for dominant and co-dominant markers are presented. These different marker types can be integrated into a single regression model by means of simple kernel operations. Field trial data from the grain maize breeding programme of the private company RAGT R2n are used to assess the predictive capabilities of the proposed methodology. Prediction accuracies are compared to those of one of today’s best performing prediction methods based on best linear unbiased prediction. Results on our data indicate that both methods match each other’s prediction accuracies for several combinations of marker types and traits. The ɛ-SVR framework, however, allows for a greater flexibility in combining different kinds of predictor variables.     

Purple non-sulphur bacteria and plant production: benefits for fertilization,stress resistance and the environment

Purple non-sulphur bacteria (PNSB) are phototrophic microorganisms, which increasingly gain attention in plant production due to their ability to produce and accumulate high-value compounds that are beneficial for plant growth. Remarkable features of PNSB include the accumulation of polyphosphate, the production of pigments and vitamins and the production of plant growth-promoting substances (PGPSs). Scattered case studies on the application of PNSB for plant cultivation have been reported for decades, yet a comprehensive overview is lacking. This review highlights the potential of using PNSB in plant production, with emphasis on three key performance indicators (KPIs): fertilization, resistance to stress (biotic and abiotic) and environmental benefits. PNSB have the potential to enhance plant growth performance, increase the yield and quality of edible plant biomass, boost the resistance to environmental stresses, bioremediate heavy metals and mitigate greenhouse gas emissions. Here, the mechanisms responsible for these attributes are discussed. A distinction is made between the use of living and dead PNSB cells, where critical interpretation of existing literature revealed the better performance of living cells. Finally, this review presents research gaps that remain yet to be elucidated and proposes a roadmap for future research and implementation paving the way for a more sustainable crop production.     

Phosphorus mining efficiency declines with decreasing soil P concentration and varies across crop species

High soil P concentrations hinder ecological restoration of biological communities typical for nutrient-poor soils. Phosphorus mining, i.e., growing crops with fertilization other than P, might reduce soil P concentrations. However, crop species have different P-uptake rates and can affect subsequent P removal in crop rotation, both of which may also vary with soil P concentration. In a pot experiment with three soil-P-levels (High-P: 125–155 mg P_Olsen/kg; Mid-P: 51–70 mg P_Olsen/kg; Low-P: 6–21 mg P_Olsen/kg), we measured how much P was removed by five crop species (buckwheat, maize, sunflower, flax, and triticale). Total P removal decreased with soil-P-level and depended upon crop identity. Buckwheat and maize removed most P from High-P and Mid-P soils and triticale removed less P than buckwheat, maize, and sunflower at every soil-P-level. The difference in P removal between crops was, however, almost absent in Low-P soils. Absolute and relative P removal with seeds depended upon crop species and, for maize and triticale, also upon soil-P-level. None of the previously grown crop species significantly affected P removal by the follow-up crop (perennial ryegrass). We can conclude that for maximizing P removal, buckwheat or maize could be grown.     

Effect of the mycotoxin deoxynivalenol on grain aphid <Emphasis Type="Italic">Sitobion avenae</Emphasis> and its parasitic wasp <Emphasis Type="Italic">Aphidius ervi</Emphasis> through food chain contamination

We investigated the influence of the Fusarium mycotoxin deoxynivalenol on the English grain aphid Sitobion avenae (Hemiptera: Aphididae) and its parasitic wasp Aphidius ervi (Hymenoptera: Braconidae) using in vitro laboratory experiments. In this tritrophic interaction, deoxynivalenol caused lethal (declined survival) and sublethal (prolonged nymphal development and reduced reproduction) effects on S. avenae aphids and consequentially led to a decreased production of parasitoid offspring resulting from parasitized deoxynivalenol-contaminated aphids. This paper highlights that the presence of mycotoxins should be considered in environmental risk assessment tests because they may alter the efficiency of biological control agents such as parasitoids through food chain contamination.     

Study of the aggressiveness of Rhizoctonia solani isolates

This paper presents an in vitro test to screen the pathogenicity of different Rhizoctonia solani isolates on a host range. The level of aggressivity of the different isolates was different for several host plants tested. There were significant differences between the crops and the isolates tested. In general, the disease level was higher on beans, lettuce and cabbage. In carrot and rye grass the level of infection was lower for the isolates of R. solani tested. The potato isolates of R. solani were less aggressive than the isolates coming from maize, fodder beet and sugar beet. The R. solani isolates were also biochemically characterized by pectic zymograms: the isolates Rs0401 (from maize) and Rs0504 (from sugar beet) belong both to the anastomosis group AG2-2.     

Clade classification of monolignol biosynthesis gene family members reveals target genes to decrease lignin in Lolium perenne

In monocots, lignin content has a strong impact on the digestibility of the cell wall fraction. Engineering lignin biosynthesis requires a profound knowledge of the role of paralogues in the multigene families that constitute the monolignol biosynthesis pathway. We applied a bioinformatics approach for genome‐wide identification of candidate genes in Lolium perenne that are likely to be involved in the biosynthesis of monolignols. More specifically, we performed functional subtyping of phylogenetic clades in four multigene families: 4CL, COMT, CAD and CCR. Essential residues were considered for functional clade delineation within these families. This classification was complemented with previously published experimental evidence on gene expression, gene function and enzymatic activity in closely related crops and model species. This allowed us to assign functions to novel identified L. perenne genes, and to assess functional redundancy among paralogues. We found that two 4CL paralogues, two COMT paralogues, three CCR paralogues and one CAD gene are prime targets for genetic studies to engineer developmentally regulated lignin in this species. Based on the delineation of sequence conservation between paralogues and a first analysis of allelic diversity, we discuss possibilities to further study the roles of these paralogues in lignin biosynthesis, including expression analysis, reverse genetics and forward genetics, such as association mapping. We propose criteria to prioritise paralogues within multigene families and certain SNPs within these genes for developing genotyping assays or increasing power in association mapping studies. Although L. perenne was the target of the analyses presented here, this functional subtyping of phylogenetic clades represents a valuable tool for studies investigating monolignol biosynthesis genes in other monocot species.     

Automatic wheat ear counting using machine learning based on RGB UAV imagery

In wheat (Triticum aestivum L) and other cereals, the number of ears per unit area is one of the main yield‐determining components. An automatic evaluation of this parameter may contribute to the advance of wheat phenotyping and monitoring. There is no standard protocol for wheat ear counting in the field, and moreover it is time consuming. An automatic ear‐counting system is proposed using machine learning techniques based on RGB (red, green, blue) images acquired from an unmanned aerial vehicle (UAV). Evaluation was performed on a set of 12 winter wheat cultivars with three nitrogen treatments during the 2017–2018 crop season. The automatic system uses a frequency filter, segmentation and feature extraction, with different classification techniques, to discriminate wheat ears in micro‐plot images. The relationship between the image‐based manual counting and the algorithm counting exhibited high levels of accuracy and efficiency. In addition, manual ear counting was conducted in the field for secondary validation. The correlations between the automatic and the manual in‐situ ear counting with grain yield were also compared. Correlations between the automatic ear counting and grain yield were stronger than those between manual in‐situ counting and GY, particularly for the lower nitrogen treatment. Methodological requirements and limitations are discussed.     

Priming of Wheat with the Green Leaf Volatile Z-3-Hexenyl Acetate Enhances Defense against Fusarium graminearum But Boosts Deoxynivalenol Production

Priming refers to a mechanism whereby plants are sensitized to respond faster and/or more strongly to future pathogen attack. Here, we demonstrate that preexposure to the green leaf volatile Z-3-hexenyl acetate (Z-3-HAC) primed wheat (Triticum aestivum) for enhanced defense against subsequent infection with the hemibiotrophic fungus Fusarium graminearum. Bioassays showed that, after priming with Z-3-HAC, wheat ears accumulated up to 40% fewer necrotic spikelets. Furthermore, leaves of seedlings showed significantly smaller necrotic lesions compared with nonprimed plants, coinciding with strongly reduced fungal growth in planta. Additionally, we found that F. graminearum produced more deoxynivalenol, a mycotoxin, in the primed treatment. Expression analysis of salicylic acid (SA) and jasmonic acid (JA) biosynthesis genes and exogenous methyl salicylate and methyl jasmonate applications showed that plant defense against F. graminearum is sequentially regulated by SA and JA during the early and later stages of infection, respectively. Interestingly, analysis of the effect of Z-3-HAC pretreatment on SA- and JA-responsive gene expression in hormone-treated and pathogen-inoculated seedlings revealed that Z-3-HAC boosts JA-dependent defenses during the necrotrophic infection stage of F. graminearum but suppresses SA-regulated defense during its biotrophic phase. Together, these findings highlight the importance of temporally separated hormone changes in molding plant health and disease and support a scenario whereby the green leaf volatile Z-3-HAC protects wheat against Fusarium head blight by priming for enhanced JA-dependent defenses during the necrotrophic stages of infection.Biogenic volatile organic compounds (BVOCs) are known regulators of communication of sedentary plants with their direct environment (Dudareva et al., 2006). Besides attracting pollinators (Pichersky and Gershenzon, 2002), repelling insect herbivores (Birkett et al., 2010), and exerting direct antimicrobial properties (Friedman et al., 2002), BVOCs can act as an alarm signal to warn neighboring plants of an imminent herbivorous or pathogen attack (Heil and Ton, 2008) or serve as an intraplant signal for the induction of resistance (Karban et al., 2006). Engelberth et al. (2004) found that maize (Zea mays) seedlings emitted the green leaf volatiles (GLVs) Z-3-hexenal, Z-3-hexenol (Z-3-HOL), and Z-3-hexenyl acetate (Z-3-HAC) after they had been infested with caterpillars of Spodoptera exigua. Neighboring uninfested seedlings that had been exposed to these GLVs subsequently showed a considerable higher production of the plant defense hormone jasmonic acid (JA) after treatment with caterpillar regurgitant. This form of induced resistance is called priming. Plants in a primed state display faster and/or stronger activation of defense pathways when challenged by microbial pathogens, herbivorous insects, or abiotic stresses (Conrath, 2009). Exposure to these priming signals does not entail a direct activation of costly defense mechanisms but rather a stronger up-regulation of defense pathways when the plant is actually under attack (van Hulten et al., 2006). Besides resulting in a stronger induction of the JA pathway, priming also has been shown to enhance defense associated with the salicylic acid (SA) pathway, which plays a critical role in plant defense against biotrophic pathogens (Conrath et al., 2006; Jung et al., 2009).The lion’s share of attention on the use of GLVs in induced resistance has been directed to plant-insect interactions. However, the literature regarding priming by GLVs in plant-pathogen interactions remains scarce (Heil, 2014). Few studies have been performed investigating the effect of priming by GLVs on plant-fungus interactions (Scala et al., 2013a, and refs. therein). For example, hexanoic acid, a molecule with a similar structure to GLVs, has been shown to act as a priming agent in tomato (Solanum lycopersicum) plants against an infection by the necrotrophic fungus Botrytis cinerea, leading to a reduced accumulation of reactive oxygen species in primed plants (Vicedo et al., 2009; Kravchuk et al., 2011; Finiti et al., 2014). Since the GLVs E-2-hexenal (E-2-HAL), Z-3-HOL, E-2-hexenol, and Z-3-HAC also have been reported to be emitted by perennial ryegrass (Lolium perenne) after infection with Fusarium poae (Pańka et al., 2013) and by wheat (Triticum aestivum) seedlings after infection with Fusarium graminearum (Piesik et al., 2011), one may speculate that GLVs not only serve as a priming agent against the impending threat of herbivorous insects but rather constitute a general warning and priming mechanism against insects, bacteria, and fungi alike.Fusarium head blight (FHB) is an important disease in cereals caused by a complex of Fusarium spp., of which the hemibiotroph F. graminearum is one of the most prevalent (Parry et al., 1995; Goswami and Kistler, 2004; Audenaert et al., 2009). Besides yield losses of up to 40%, FHB also confers quality losses because of the production of mycotoxins such as deoxynivalenol (DON; Parry et al., 1995; Bottalico and Perrone, 2002; Vanheule et al., 2014).The hemiobiotrophic nature of F. graminearum entails that its lifestyle is characterized by a biotrophic phase followed by a necrotrophic phase. During the biotrophic phase, spores will germinate and hyphae will grow extracellularly and intercellularly. To counteract fungal colonization during the biotrophic phase, the host plant will accumulate hydrogen peroxide (H₂O₂) to induce programmed cell death. However, H₂O₂ acts as a signal for F. graminearum to produce DON, which in turn creates a positive feedback loop leading to increased H₂O₂ and DON production, clearing the way for F. graminearum to further colonize the host plant (Desmond et al., 2008). Plant defense against the biotrophic and necrotrophic phases generally has been linked to SA- and JA-related pathways, respectively (Glazebrook, 2005). This was also found in the study by Ding et al. (2011). They reported higher endogenous SA concentrations during the first hours of infection, followed by a rise in JA concentrations later on. However, plant defense against pathogens is regulated by a whole array of plant hormones, between which an intricate cross talk exists (Pieterse et al., 2012). One of the best-studied antagonistic signaling pathways is between SA and JA (Thaler et al., 2002; Pieterse et al., 2012). Research investigating the hormonal modulation of plant immunity has been done primarily in dicots. The negative relationship between SA and JA also seems to be conserved in rice (Oryza sativa), another monocot (De Vleesschauwer et al., 2013). Because of the presence of this possible antagonistic signaling and the hemibiotrophic lifestyle of F. graminearum, it is important to look more closely to the effect of priming on these two defense pathways in wheat.Here, we show that preexposure of wheat to the GLV Z-3-HAC primes wheat plants for an enhanced defense against a future infection with F. graminearum. Furthermore, our results indicate that pretreatment with Z-3-HAC leads to a stronger activation of JA-related defense while exerting suppressive effects on SA-responsive gene expression. Lastly, we found evidence that enhanced plant defense led to increased DON production by F. graminearum.