Cercospora leaf blight of soybean (Glycine max) is a serious problem in the mid-south of the United States and is present in most soybean-growing regions of the world. The causal organisms, Cercospora kikuchii, C. cf. flagellaris and C. cf. sigesbeckiae, utilize the photo-activated toxin cercosporin as the primary pathogenicity factor. The disease has historically been understood to progress from a purpling or bronzing of the leaves to a blight in which tissue dies. Previous literature showed possible binding of cercosporin to metals and previous work in our group showed possible use of minor element nutrition in the plant to reduce disease severity. Through the use of UV-vis spectrometry, nuclear magnetic resonance spectroscopy, and mass spectrometry evidence is given in support of cercosporin-metal binding. From this work, it is hypothesized that the metals bind at the phenol group of the cercosporin molecule, yielding a dark purple compound. Cercosporin appears to have higher affinity for iron, especially the Fe(III) form, over the other plant minor elements but appears to have some minor binding affinity for aluminum, copper, manganese, and zinc. Cercosporin and iron combinations show greatly reduced toxicity compared to cercosporin alone, in support of previous reports in the literature. Manganese added to cercosporin resulted in more necrosis of the soybean tissue. Soybean plants fertilized with iron sufficient to increase tissue concentrations by 20 mg/kg dry weight resulted in a significantly higher resistance to the toxin. The iron fertilization also resulted in a roughly 60 mg/kg dry weight reduction in manganese, which may also help account for the reduction in toxicity. The symptoms also were less similar to the typical blight symptoms and more similar to the purple leaf symptoms. Isolates of C. cf. flagellaris grown on split chrome azurol sulfate (CAS) plates elicited the signature color shift indicative of siderophore production. Cultures of C. cf. flagellaris also were grown in liquid minimal medium amended with cercosporin, iron, or both. In broths containing both cercosporin and iron, fungal growth was higher, total cercosporin production was lower, and iron content of the mycelium was significantly higher than that of controls. This demonstrates that the fungus is capable of up-taking and incorporating iron in this form. When only cercosporin was provided, mycelial concentrations of aluminum, copper, and boron were higher than other treatments. This indicates that cercosporin may be involved in other metal uptake and these metals may compete with iron. While previous work indicated cercosporins ability to bind iron, none gave conclusive evidence, analyzed the mechanisms, or investigated the purpose. Findings presented herein suggest that cercosporin is capable of acting as a siderophore; serving both as the primary pathogenicity factor and as an iron acquisition mechanism. Furthermore, the role of iron in disease development and symptom progression should be re-evaluated along with its effects on yield. The symptoms of purpling and blight appear to be correlated with the amount of iron in soybean tissue. Further work should investigate ways to practically manipulate these results to lower damage caused by these pathogens.

To study the effect of Silicon (Si) with and without a fungicide on anthracnose development of sorghum, several experiments were conducted in the greenhouse and field. In the initial study, different Si rates [0 (control), 0 (lime control), 200, 400, 600, 800 kg Si/ha] were used in a low-Si Alfisol to determine if inoculum densities were affected by Si levels. No differences were observed between inoculum densities of 1*105 and 1*106 conidia/ml in affecting anthracnose development in the greenhouse. Anthracnose severity was found to be lowest in plants treated with 800 kg Si/ha, regardless of inoculum density. In another study, the effect of Si on moderately susceptible (Pioneer 84G62) and moderately resistant (Pioneer 84P80) hybrids was also examined with lower inoculum concentration under greenhouse conditions. Fungicide (Pyraclostrobin) was also included to suppress the anthracnose development. Silicon had a significant effect on plant Si concentration and anthracnose development. Anthracnose severity was reduced as plant and soil Si levels increased. The highest Si application rate (800 kg Si/ha) reduced Final Disease Severity (FDS) and Area Under Anthracnose Progress Curve (AUAPC) by 18 and 36% as compared to the control for the first greenhouse experiment (p<0.05). Likewise, it reduced FDS and AUAPC of the 2nd greenhouse experiment by 76 and 67%, respectively (p<0.001). Pyraclostrobin effectively reduced AUAPC by 50 and 36%, respectively, for the two greenhouse experiments. Similar Si + pyraclostrobin experiments were conducted under field conditions at Dean Lee (Inceptisols) and Winnsboro (Alfisols), Louisiana with higher soil Si levels (120 µg/g and 40 µg/g respectively). Even though soil Si increased with higher Si rates for both fields, no significant increase in Si accumulation in sorghum leaves or grains was observed. At Dean Lee, pyraclostrobin reduced AUAPC by 44 and 39% and FDS by 50 and 48%; respectively, for Pioneer 84G62 and Pioneer 84P80 (p<0.001). However, pyraclostrobin had no effect in reducing anthracnose at Winnsboro. Yield was higher for Pioneer 84G62 than Pioneer 84P80 at Dean Lee. Silicon had a greater impact in suppressing anthracnose development on low-Si soils under greenhouse conditions than high-Si soils of the field.

Kabuli chickpea (<i>Cicer arietinum</i> L.) has two leaf types, the fern and unifoliate. Yield potential is limited for kabuli chickpea in Saskatchewan. It is limited by a short-season, a semi-arid environment, and end-of-season rainfall. Manipulating plant population, and choosing chickpea cultivars with the best leaf type for biomass production, radiation interception and yield for the early, middle, or late growth season, may increase chickpea yield. Therefore, the objectives of this study were: to (i) determine the relationship between leaf type and key growth parameters of six chickpea cultivars varying in leaf morphology at moderate and high plant population densities; (іі) to characterize the reaction of the fern and unifoliate leaf to altered canopy light environments. Different light environments were created by 50% defoliation at vegetative growth, first flower, and 50% shading from vegetative growth to first flower, as well as two light enrichment treatments initiated at the first flower and pod formation stages.<p> Fern leaf cultivars exhibited higher maximum light interception, seasonal cumulative intercepted radiation and a higher harvest index compared to unifoliate leaf cultivars. However, both leaf type canopies had less than 95% light interception for most of the season. The fern and unifoliate leaf type contributed to similar radiation use efficiency in three out of four location-years. In addition, fern leaf cultivars produced significantly higher seed yield than cultivars with unifoliate leaves.<p>Plant density influenced growth parameters. For example, the 45 plants m-2 treatment had a higher harvest index than the 85 plants m-2 treatment, in two location-years, while both population treatments were similar in the other two location-years. Yield of chickpea was increased by higher plant population in only one location-year, but was not significantly affected by plant population in the other location-years. The effect of canopy light environment manipulation on chickpea yield depended on the stages of plant development when they were applied. Defoliation at vegetative growth and first flower had no effect on yield. However, plants responded significantly to the 50% shade treatment; the crop growth rate, harvest index and yield were less in the shaded treatment compared to the control. Shading also increased plant height. Light enrichment treatments increased the yield. However, the degree of yield increase was greater when light enrichment occurred at first flower, than at the later stage of pod formation. These results highlighted the importance of the amount of irradiance during the flowering stage. It was concluded that chickpea breeders should select lines with fern leaves for improved radiation interception when breeding cultivars for semiarid short-season environments such as in Saskatchewan. Management and breeding practices should ensure that the crop can make efficient use of the solar radiation at flowering to maximize yield. Improvement at the canopy and subsequent yield level is yet to be made in Saskatchewan environments by increased light interception, increased growth before flowering, and increased and stable harvest index.

Kabuli chickpea (<i>Cicer arietinum</i> L.) has two leaf types, the fern and unifoliate. Yield potential is limited for kabuli chickpea in Saskatchewan. It is limited by a short-season, a semi-arid environment, and end-of-season rainfall. Manipulating plant population, and choosing chickpea cultivars with the best leaf type for biomass production, radiation interception and yield for the early, middle, or late growth season, may increase chickpea yield. Therefore, the objectives of this study were: to (i) determine the relationship between leaf type and key growth parameters of six chickpea cultivars varying in leaf morphology at moderate and high plant population densities; (іі) to characterize the reaction of the fern and unifoliate leaf to altered canopy light environments. Different light environments were created by 50% defoliation at vegetative growth, first flower, and 50% shading from vegetative growth to first flower, as well as two light enrichment treatments initiated at the first flower and pod formation stages.<p> Fern leaf cultivars exhibited higher maximum light interception, seasonal cumulative intercepted radiation and a higher harvest index compared to unifoliate leaf cultivars. However, both leaf type canopies had less than 95% light interception for most of the season. The fern and unifoliate leaf type contributed to similar radiation use efficiency in three out of four location-years. In addition, fern leaf cultivars produced significantly higher seed yield than cultivars with unifoliate leaves.<p>Plant density influenced growth parameters. For example, the 45 plants m-2 treatment had a higher harvest index than the 85 plants m-2 treatment, in two location-years, while both population treatments were similar in the other two location-years. Yield of chickpea was increased by higher plant population in only one location-year, but was not significantly affected by plant population in the other location-years. The effect of canopy light environment manipulation on chickpea yield depended on the stages of plant development when they were applied. Defoliation at vegetative growth and first flower had no effect on yield. However, plants responded significantly to the 50% shade treatment; the crop growth rate, harvest index and yield were less in the shaded treatment compared to the control. Shading also increased plant height. Light enrichment treatments increased the yield. However, the degree of yield increase was greater when light enrichment occurred at first flower, than at the later stage of pod formation. These results highlighted the importance of the amount of irradiance during the flowering stage. It was concluded that chickpea breeders should select lines with fern leaves for improved radiation interception when breeding cultivars for semiarid short-season environments such as in Saskatchewan. Management and breeding practices should ensure that the crop can make efficient use of the solar radiation at flowering to maximize yield. Improvement at the canopy and subsequent yield level is yet to be made in Saskatchewan environments by increased light interception, increased growth before flowering, and increased and stable harvest index.

Microalgae are a rich, largely untapped source of lipids. Algae are underutilized, in part, because lipid formation generally is stimulated by stress, such as nutrient deficiency. Nutrient deficiencies reduce growth, resulting in a tradeoff between elevated cellular lipids and abundant cell division. This tradeoff is not well understood. We also have a poor understanding of the physiological drivers for this lipid formation. Here we report on three sets of research: 1) Assessment of species differences in growth and lipid content tradeoffs with high and low level nitrogen deficiency; 2) Investigation of physiological drivers of lipid formation, by mass balance accounting of cellular nitrogen with progressing deficiency; 3) Examination of the effects of sodium chloride and silicon on lipid production in a marine diatom. 1) Nitrogen deficiency typically had disproportionate effects on growth and lipid content, with profound differences among species. Optimally balancing the tradeoff required a wide range in the rate of nitrogen supply to species. Some species grew first and then accumulated lipids, while other species grew and accumulated lipids concurrently--a characteristic that increased lipid productivity. High lipid content generally resulted from a response to minimal stress. 2) Commonalities among species in cellular nitrogen at the initiation of lipid accumulation provided insight into the physiological drivers for lipid accumulation in nitrogen deficient algae. Total nitrogen uptake and retention differed widely among species, but the ratio of minimum retained nitrogen to nitrogen at the initiation of lipid accumulation was consistent among species at 0.5 ± 0.04. This suggests that lipid accumulation was signaled by a common magnitude of nitrogen deficiency. Among the cellular pools of nitrogen at the initiation of lipid accumulation, the concentration of RNA and the protein to RNA ratio were most similar among species with averages of 3.2 ± 0.26 g L-1 (8.2% variation) and 16 ± 1.5 (9.2% variation), respectively. This implicates critical levels of these parameters as potential signals initiating the accumulation of lipids. 3) In a marine diatom, low levels of either sodium chloride or silicon resulted in at least 50% increases in lipid content. The synergy of simultaneous, moderate sodium chloride and silicon stress resulted in lipid content up to 73%. There was a strong sodium chloride/silicon interaction in total and ash-free dry mass densities that arose because low sodium chloride was inhibitory to growth, but the inhibition was overcome with excessive silicon supply. This suggests that low sodium chloride may have affected metabolism of silicon.

<p>Nitrogen (N) is one of the major nutrient elements that affects growth, development, and quality of floriculture crops. Both sub-optimal and supra-optimal levels of N can negatively affect crop growth. In addition, over- fertilization may cause run-off and leaching of the N fertilizer leading to environmental pollution. Therefore, it is crucial to maintain optimal N level in plant tissue to produce good quality crops and increase productivity. This requires regular monitoring and measurement of plant N status. Laboratory analysis, the only direct method available to measure tissue N content, is destructive of plant tissue and expensive. Other available indirect methods are laborious, expensive, and/ or less reliable. In addition to measuring plant N status, it is crucial to understand acclimation responses at biochemical, leaf, and whole-plant levels in floriculture crops to N-deficit conditions. This will aid in developing a mechanistic model of plant responses to sub-optimal levels of N, proper fertilizer guidelines during production, and screening tools for identifying new varieties with tolerance to low-N level in the root zone. Unfortunately, there is limited research on floriculture crops that is simultaneously focused on plant responses at different scales to N-deficit conditions. The objectives of this research were to (i) assess the feasibility of image-based reflectance ratios for estimating tissue N content in poinsettia (Expt. 1), (ii) develop an affordable, remote sensor that can accurately and non-destructively estimate tissue N <a>content</a> in poinsettia (Expt. 2), (iii) study the physiological acclimation at whole-plant, leaf, and biochemical scales in poinsettia cultivars to N-deficit conditions (Expt. 3).</p> <p>In Expt. 1, we compared several spectral ratios based on the ratio of reflectance of near infrared <i>(R₈₇₀)</i> to reflectance of blue (<i>R</i><i>₈₇₀/R₄₅₀</i>), green (<i>R₈₇₀/R₅₂₁</i>), yellow (<i>R₈₇₀/R₅₉₃</i>), red (<i>R₈₇₀/R₆₂₅</i>), hyper-red (<i>R₈₇₀/R₆₆₀</i>), and far-red(<i>R₈₇₀/R₇₃₀</i>) wavelengths from plants<i></i>to measure whole-plant tissue N content in<i></i>four cultivars of poinsettia (<i>Euphorbia pulcherrima</i>) using a multispectral image station. Results indicated the reflectance ratio <i>R₈₇₀/R₆₂₅</i> was most suitable for assessing tissue N content in plants. In Expt. 2, a low-cost remote sensor was developed based on the findings of Expt. 1 that captured red and near-infrared images of plants, from which a reflectance ratio (<i>R_ratio</i>) was developed. The ratio was linearly related to tissue N content in all poinsettia cultivars. Furthermore, <i>R_ratio</i>was found to be more specific to N than to other elements in the tissue and related to the chlorophyll concentration of the plant. In Expt. 3, poinsettia cultivars ‘Jubilee Red’ (‘JR’) and ‘Peterstar Red’ (‘PSR’) displayed different acclimation strategies for physiology and growth under N-deficit conditions. Significantly higher growth was observed in ‘JR’ than in ‘PSR’ in the sub-optimal treatment, which indicates that ‘JR’ is more tolerant to N stress compared to ‘PSR’. Further analyses indicated that N uptake was higher in ‘JR’ than in ‘PSR’ under N-deficit conditions, without any changes in root morphology or growth. This is possible when higher levels of energy are available to transport nitrate and/or ammonia from the substrate into the root cells. Supporting this, significantly higher photosynthesis and carboxylation efficiency were observed in ‘JR’ than ‘PSR’ under N-deficit condition. These results shows that higher growth of ‘JR’ than ‘PSR’ under N-deficit conditions was likely due to increased N uptake (likely due to increased energy-driven transporter activity), which increased tissue N and chlorophyll levels. Further, these increases resulted in higher carboxylation efficiency and photosynthesis by ‘JR’ than ‘PSR’. Increased carbohydrate synthesis supported leaf growth and provided required energy in the fine root cells for N uptake from the substrate.</p>

No description available.

<p>Dissecting the soybean grain yield (GY) to approach it as a sum of its associated processes seems a viable approach to explore this trait considering its complex multigenic nature. Monteith (1972, 1977) first defined potential yield as the result of three physiological efficiencies: light interception (Ei), radiation use efficiency (RUE) and harvest index (HI). Though this rationality is not recent, few works assessing these three efficiencies as strategies to improve crops have been carried out. This thesis approaches yield from the perspective of Ei, RUE, and HI to better understand yield as the result of genetic and physiological processes. This study reveals the phenotypic variation, heritability, genetic architecture, and genetic relationships for Ei, RUE, and HI and their relationships with GY and other physiological and phenological variables. Similarly, genomic prediction is presented as a viable strategy to partially overcome the tedious phenotyping of these traits. A large panel of 383 soybean recombinant inbred lines (RIL) with significant yield variation but shrinkage maturity was evaluated in three field environments. Ground measurements of dry matter, photosynthesis (A), transpiration (E), water use efficiency (WUE), stomatal conductance (gs), leaf area index (LAI) and phenology (R1, R5, R8) were measured. Likewise, RGB imagery from an unmanned aircraft system (UAS) were collected with high frequency (~12 days) to estimate the canopy dynamic through the canopy coverage (CC). Light interception was modeled through a logistic curve using CC as a proxy and later compared with the seasonal cumulative solar radiation collected from weather stations to calculate Ei. The total above ground biomass collected during the growing season and its respective cumulative light intercepted were used to derive RUE through linear models fitting, while apparent HI was calculated through the ratio seeds dry matter vs total above-ground dry matter. Additive-genetic correlations, genome wide association (GWA) and whole genome regressions (WGR) were performed to determine the relationship between traits, their association with genomic regions, and the feasibility of predicting these efficiencies through genomic information. Our results revealed moderate to high phenotypic variation for Ei, RUE, and HI. Additive-genetic correlation showed a strong relationship of GY with HI and moderate with RUE and Ei when the whole data set was considered, but negligible contribution of HI on GY when just the top 100 yielding RILs were analyzed. High genetic correlation to grain yield (GY) was also observed for A (0.87) and E (0.67), suggesting increase in GY can be achieved through the improvement of A or E. The GWA analyses showed that Ei is associated with three SNPs; two of them located on chromosome 7 and one on chromosome 11 with no previous quantitative trait loci (QTLs) reported for these regions. RUE is associated with four SNPs on chromosomes 1, 7, 11, and 18. Some of these QTLs are novel, while others are previously documented for plant architecture and chlorophyll content. Two SNPs positioned on chromosome 13 and 15 with previous QTLs reported for plant height and seed set, weight and abortion were associated with HI. WGR showed high predictive ability for Ei, RUE, and HI with maximum correlation ranging between 0.75 to 0.80. Both directed and undirected multivariate explanatory models indicate that HI has a strong relationship with A, average growth rate of canopy coverage for the first 40 days after planting (AGR40), seed-filling (SFL), and reproductive length (RL). According to the path analysis, increase in one standard unit of HI promotes changes in 0.5 standard units of GY, while changes in the same standard unit of RUE, and Ei produce increases on GY of 0.20 and 0.19 standard units. This study presents novel genetic knowledge for Ei, RUE, HI and GY along with a set of tools that may contribute to the development of new cultivars with enhanced light interception, light conversion and optimized dry matter partitioning in soybean. This work not only complements the physiological knowledge already available with the genetic control of traits directly associated with yield, but also represents a pioneer attempt to integrate traditional physiological traits into the breeding process in the context of physiological breeding<br></p>

La demande mondiale de nourriture et d’energie est en constante augmentation. L’accès à de nouvelles terres arables étant limité, les agriculteurs devront produire environ 70% de plus par hectare d'ici 2050. L'optimisation et la rationalisation des systèmes de production agricole sont essentielles pour assurer la sécurité alimentaire des populations dans des conditions durables. La communauté agricole serait en mesure d'optimiser les performances des cultures en ayant accès à une meilleure compréhension des systèmes de production, notamment concernant leur potentiel et principaux facteurs limitants. Dans ce contexte, la caractérisation des conditions de croissance des cultures prenant en compte leurs impacts sur le potentiel de production et sur les pertes de rendement est importante pour identifier les marges de progrès et proposer des systèmes agricoles améliorés.Dans cette étude une approche est proposée pour caractériser l'environnement de grandes zones géographiques prenant en compte les impacts des ressources clés au développement des plantes (température, rayonnement, eau et azote) sur la performance des cultures. Le niveau de disponibilité de ces ressources a été estimé en utilisant des approches de modélisation de culture, intégrant données météorologiques et caractéristiques de sol. L'efficacité d'utilisation des ressources disponibles a été caractérisée par les pratiques culturales ainsi que la stratégie et l’environnement technique des agriculteurs. Une procédure générique d’intégration de la donnée a été établie et utilisée pour décrire la production de maïs grain en Amérique du Nord de 1991 à 2013. Le jeu de données résultant de cette intégration couvre 84% des superficies plantées en maïs, décrites par 1 558 unités spatiales élémentaires, pour un total de 28 303 scénarios indépendants. Cette méthode combine et harmonise les observations de rendement issues des statistiques agricoles avec un large ensemble de descripteurs pertinents de conditions de croissance.Un sous-ensemble de 21 facteurs limitants a été identifié par une analyse de sélection de variables pour expliquer 66% de la variabilité des pertes de rendement observées. Compte tenu du nombre de dimensions du jeu de donnée, la méthode relaxed lasso a été choisie comme un compromis intéressant entre capacité de prédiction et d’interprétation. Les descripteurs sélectionnés montrent la contribution de la gestion des cultures à la variabilité des pertes de rendement, en particulier les niveaux d'intensification de l’agriculture et l’environnement technique des agriculteurs. Le troisième facteur principal est la disponibilité en eau et la sécheresse qui en résulte. L’utilisation des connaissances en physiologie dans la conception de descripteurs a considérablement amélioré la capacité d’interprétation de la proposition et la confiance des utilisateurs finaux dans l'approche.La structure des facteurs limitant le rendement a été utilisée pour identifier les Environnements Types les plus fréquents. Deux approches ont été proposes : l’une axée uniquement sur les facteurs physiologiques et l’autre intégrant également l’identification de marchés de taille homogène. 11 Environnements Types ont été identifiés en fonction des facteurs physiologiques et 8 en combinant facteurs physiologiques et contraintes de taille de marché. Les Environnements Types basés sur les facteurs physiologiques sont très informatifs sur l'évolution historique des pratiques culturales et les changements de stratégie des agriculteurs au cours du temps. Une telle catégorisation des conditions de croissance a montré des capacités à reproduire l'expertise de terrain et à soutenir l'évaluation de génotypes. Trois cas d'utilisation ont servi à illustrer l'intérêt de l'approche pour (i) décrire l'historique environnemental d'un marché, (ii) définir des populations d’Environnements cibles (TPE) et (iii) évaluer la pertinence de l'échantillonnage environnemental de réseaux multi-lieux (MET). / The global demand for food and energy is constantly increasing. As access to spare arable land is limited, growers will have to produce about 70% more per hectare by 2050. Optimization and rationalization of the agricultural production systems is then critical to ensure food security under sustainable conditions. The agricultural community would be in a position to optimize crop performance by better understanding the cropping systems, especially regarding potential production and limiting factors. In this context, the ability to characterize crop growing conditions in regards to their impacts on potential yield and yield gap is important in order to identify margins of progress and propose improved farming systems.An approach is proposed to characterize the crop environment of large geographical areas taking into consideration the impact on crop performance of the key resources for plant development (temperature, solar radiation, water and nitrogen). The level of availability of each resource was estimated by using crop modeling approaches integrating weather data and soil characteristics as inputs. The efficiency of use of the available resources was characterized by crop practices, grower strategy and grower technical environment. A generic integration procedure was established and used to describe corn production for grain in North America from 1991 to 2013. The resulting dataset covers 84% of the total corn planted area, deciphered in 1,558 elementary spatial units, for a total of 28,303 independent scenarios. Such a method combines and harmonizes, at scale, yield observations from agricultural statistics with a large set of relevant descriptors of growing conditions.A subset of 21 main limiting factors was identified through variable-selection analysis to explain 66% of the observed yield-gap variability. The relaxed lasso method resulted in an interesting compromise between interpretability and prediction ability. The selected descriptors highlighted the contribution of crop management in yield-gap variability, especially regarding levels of crop intensification and the technical environment of growers. The third main factor is water availability and resulting drought. The integration of knowledge in crop physiology into descriptor design significantly improved the interpretability of the proposal and the confidence of end-users in the approach.Yield-gap patterns were used to identify the set of most frequent Environment Types. Two approaches were proposed: one focused on crop physiology drivers and another also considered market-size homogeneity among Environment Types. This resulted in 11 Environment Types based on crop physiology drivers and 8 Environment Types when combining crop physiology drivers and market constraints. Crop-physiology Environment Types are very informative on the historical evolution of cultural practices and the changes in grower strategies over the studied period. Such categorization of growing conditions demonstrated the ability to reproduce field expertise and support genotype evaluation. Three business Use Cases were used to illustrate the interest of the approach in (i) describing the environmental history of a market, (ii) defining relevant Target Population Environments (TPE) and (iii) evaluating the environmental sampling relevance of Multi-Environment Trials (MET).