Assessing Grain Quality Changes in White and Black Rice under Water Deficit.

Rice is an essential diet component for a significant portion of the population worldwide. Due to the high water demand associated with rice production, improving water use efficiency and grain quality is critical to increasing the sustainability of the crop. This species includes rice varieties with diverse pigmentation patterns. Grain quality, including industrial, nutritional, and functional quality traits, of two black rice genotypes and a commercial white rice cultivar were evaluated in different locations and under different water regimes. Flooding produced higher grain weight compared to alternate wetting and drying irrigation. A high correlation was found between grain color, total phenolic content (TPC), and antioxidant activity. The black rice genotypes showed higher TPC levels and antioxidant capacity, mainly due to higher levels of cyanidin 3-O-glucoside. The phenolic profile varied between whole and polished grains, while mineral composition was influenced by location and irrigation regime. In turn, the environment influenced grain quality in terms of industrial and nutritional characteristics, with significant differences in quality between whole and polished grains. This study provides valuable information on the genotype-environment relationship in rice and its effect on grain quality, which could contribute to selecting genotypes for an appropriate environment.

Phenotypic variation in leaf photosynthetic traits, leaf area index, and carbon discrimination of field-grown wheat genotypes and their relationship with yield performance in Mediterranean environments.

MAIN CONCLUSION: Leaf water potential, gas exchange, and chlorophyll fluorescence exhibited significant differences among genotypes, high environmental effects, but low heritability. The highest-yielding and drought-tolerant genotypes presented superior harvest index and grain weight, compared to drought-susceptible ones. Physiological phenotyping can help identify useful traits related to crop performance under water-limited conditions. A set of fourteen bread wheat genotypes with contrasting grain yield (GY) was studied in eight Mediterranean environments in Chile, resulting from the combination of two sites (Cauquenes and Santa Rosa), two water conditions (rainfed-WL and irrigated-WW), and four growing seasons (2015-2018). The objectives were to (i) evaluate the phenotypic variation of leaf photosynthetic traits after heading (anthesis and grain filling) in different environments; (ii) analyze the relationship between GY and leaf photosynthetic traits and carbon isotope discrimination (Δ13C); and (iii) identify those traits that could have a greater impact in the determination of tolerant genotypes under field conditions. Agronomic traits exhibited significant genotypic differences and genotype × environment (GxE) interaction. The average GY under the WW condition at Santa Rosa was 9.2 Mg ha-1 (range 8.2-9.9 Mg ha-1) and under the WL condition at Cauquenes was 6.2 Mg ha-1 (range 3.7-8.3 Mg ha-1). The GY was closely related to the harvest index (HI) in 14 out of 16 environments, a trait exhibiting a relatively high heritability. In general terms, the leaf photosynthetic traits presented low GxE interaction, but high environmental effects and low heritability, except for the chlorophyll content. The relationships between GY and leaf photosynthetic traits were weaker when performed across genotypes in each environment, indicating low genotypic effects, and stronger when performed across environments for each genotype. The leaf area index and Δ13C also presented high environmental effects and low heritability, and their correlations with GY were influenced by environmental effects. The highest-yielding and drought-tolerant genotypes presented superior HI and grain weight, but no clear differences in leaf photosynthetic traits or Δ13C, compared to drought-susceptible ones. It seems that the phenotypic plasticity of agronomic and leaf photosynthetic traits is very important for crop adaptation to Mediterranean environments.

Exploring plant responses to abiotic stress by contrasting spectral signature changes.

In this study, daily changes over a short period and diurnal progression of spectral reflectance at the leaf level were used to identify spring wheat genotypes (Triticum aestivum L.) susceptible to adverse conditions. Four genotypes were grown in pots experiments under semi-controlled conditions in Chile and Spain. Three treatments were applied: i) control (C), ii) water stress (WS), and iii) combined water and heat shock (WS+T). Spectral reflectance, gas exchange and chlorophyll fluorescence measurements were performed on flag leaves for three consecutive days at anthesis. High canopy temperature ( H CT ) genotypes showed less variability in their mean spectral reflectance signature and chlorophyll fluorescence, which was related to weaker responses to environmental fluctuations. While low canopy temperature ( L CT ) genotypes showed greater variability. The genotypes spectral signature changes, in accordance with environmental fluctuation, were associated with variations in their stomatal conductance under both stress conditions (WS and WS+T); L CT genotypes showed an anisohydric response compared that of H CT , which was isohydric. This approach could be used in breeding programs for screening a large number of genotypes through proximal or remote sensing tools and be a novel but simple way to identify groups of genotypes with contrasting performances.

Genetic Yield Gains and Changes in Morphophysiological-Related Traits of Winter Wheat in Southern Chilean High-Yielding Environments.

Both the temperate-humid zone and the southern part of the Mediterranean climate region of Chile are characterized by high wheat productivity. Study objectives were to analyze the yield potential, yield progress, and genetic progress of the winter bread wheat (Triticum aestivum L.) cultivars and changes in agronomic and morphophysiological traits during the past 60 years. Thus, two field experiments: (a) yield potential and (b) yield genetic progress trials were conducted in high-yielding environments of central-southern Chile during the 2018/2019 and 2019/2020 seasons. In addition, yield progress was analyzed using yield historical data of a high-yielding environment from 1957 to 2017. Potential yield trials showed that, at the most favorable sites, grain yield reached ∼20.46 Mg ha-1. The prolonged growing and grain filling period, mild temperatures in December-January, ample water availability, and favorable soil conditions explain this high-potential yield. Yield progress analysis indicated that average grain yield increased from 2.70 Mg ha-1 in 1959 to 12.90 Mg ha-1 in 2017, with a 128.8 kg ha-1 per-year increase due to favorable soil and climatic conditions. For genetic progress trials, genetic gain in grain yield from 1965 to 2019 was 70.20 kg ha-1 (0.49%) per year, representing around 55% of the yield progress. Results revealed that the genetic gains in grain yield were related to increases in biomass partitioning toward reproductive organs, without significant increases in Shoot DW production. In addition, reducing trends in the NDVI, the fraction of intercepted PAR, the intercepted PAR (form emergence to heading), and the RGB-derived vegetation indices with the year of cultivar release were detected. These decreases could be due to the erectophile leaf habit, which enhanced photosynthetic activity, and thus grain yield increased. Also, senescence of bottom canopy leaves (starting from booting) could be involved by decreasing the ability of spectral and RGB-derived vegetation indices to capture the characteristics of green biomass after the booting stage. Contrary, a positive correlation was detected for intercepted PAR from heading to maturity, which could be due to a stay-green mechanism, supported by the trend of positive correlations of Chlorophyll content with the year of cultivar release.

Photosynthetic Metabolism under Stressful Growth Conditions as a Bases for Crop Breeding and Yield Improvement.

Increased periods of water shortage and higher temperatures, together with a reduction in nutrient availability, have been proposed as major factors that negatively impact plant development. Photosynthetic CO2 assimilation is the basis of crop production for animal and human food, and for this reason, it has been selected as a primary target for crop phenotyping/breeding studies. Within this context, knowledge of the mechanisms involved in the response and acclimation of photosynthetic CO2 assimilation to multiple changing environmental conditions (including nutrients, water availability, and rising temperature) is a matter of great concern for the understanding of plant behavior under stress conditions, and for the development of new strategies and tools for enhancing plant growth in the future. The current review aims to analyze, from a multi-perspective approach (ranging across breeding, gas exchange, genomics, etc.) the impact of changing environmental conditions on the performance of the photosynthetic apparatus and, consequently, plant growth.

Genotypic variations in leaf and whole-plant water use efficiencies are closely related in bread wheat genotypes under well-watered and water-limited conditions during grain filling.

Wheat plants growing under Mediterranean rain-fed conditions are exposed to water deficit, particularly during the grain filling period, and this can lead to a strong reduction in grain yield (GY). This study examines the effects of water deficit after during the grain filling period on photosynthetic and water-use efficiencies at the leaf and whole-plant level for 14 bread wheat genotypes grown in pots under glasshouse conditions. Two glasshouse experiments were conducted, one in a conventional glasshouse at the Universidad de Talca, Chile (Experiment 1), and another at the National Plant Phenomics Centre (NPPC), Aberystwyth, UK (Experiment 2), in 2015. Plants were grown under well-watered (WW) and water-limited (WL) conditions during grain filling. The reductions in leaf water potential (Ψ), net CO2 assimilation (An) and stomatal conductance (gs) due to water deficit were 79, 35 and 55%, respectively, during grain filling but no significant differences were found among genotypes. However, chlorophyll fluorescence parameters (as determined on dark-adapted and illuminated leaves) and chlorophyll content (Chl) were significantly different among genotypes, but not between water conditions. Under both water conditions, An presented a positive and linear relationship with the effective photochemical quantum yield of Photosystem II (Y(II)) and the maximum rate of electron transport (ETRmax), and negative with the quantum yield of non-photochemical energy conversion in Photosystem II (Y(NPQ)). The relationship between An and Chl was positive and linear for both water conditions, but under WL conditions An tended to be lower at any Chl value. Both, instantaneous (An/E) and intrinsic (An/gs) water-use efficiencies at the leaf level exhibited a positive and linear relationship with plant water-use efficiency (WUEp = plant dry weight/water use). Carbon discrimination (Δ13C) in kernels presented a negative relationship with WUEp, at both WW and WL conditions, and a positive relationship with GY. Our results indicate that during grain filling wheat plants face limitations to the assimilation process due to natural senesce and water stress. The reduction in An and gs after anthesis in both water conditions was mainly due a decline in the chlorophyll content (non-stomatal limitation), whereas the observed differences between water conditions were mainly due to a stomatal limitation.

Thermal Imaging Reliability for Estimating Grain Yield and Carbon Isotope Discrimination in Wheat Genotypes: Importance of the Environmental Conditions.

Canopy temperature (Tc) by thermal imaging is a useful tool to study plant water status and estimate other crop traits. This work seeks to estimate grain yield (GY) and carbon discrimination (Δ13C) from stress degree day (SDD = Tc - air temperature, Ta), considering the effect of a number of environmental variables such as the averages of the maximum vapor pressure deficit (VPDmax) and the ambient temperature (Tmax), and the soil water content (SWC). For this, a set of 384 and a subset of 16 genotypes of spring bread wheat were evaluated in two Mediterranean-climate sites under water stress (WS) and full irrigation (FI) conditions, in 2011 and 2012, and 2014 and 2015, respectively. The relationship between the GY of the 384 wheat genotypes and SDD was negative and highly significant in 2011 (r2 = 0.52 to 0.68), but not significant in 2012 (r2 = 0.03 to 0.12). Under WS, the average GY, Δ13C, and SDD of wheat genotypes growing in ten environments were more associated with changes in VPDmax and Tmax than with the SWC. Therefore, the amount of water available to the plant is not enough information to assume that a particular genotype is experiencing a stress condition.

Exploring Agronomic and Physiological Traits Associated With the Differences in Productivity Between Triticale and Bread Wheat in Mediterranean Environments.

In Mediterranean climates soil water deficit occurs mainly during the spring and summer, having a great impact on cereal productivity. While previous studies have indicated that the grain yield (GY) of triticale is usually higher than bread wheat (Triticum aestivum L.), comparatively little is known about the performance of these crops under water-limited conditions or the physiological traits involved in the different yields of both crops. For this purpose, two sets of experiments were conducted in order to compare a high yielding triticale (cv. Aguacero) and spring wheat (cvs. Pandora and Domo). The first experiment, aiming to analyze the agronomic performance, was carried out in 10 sites located across a wide range of Mediterranean and temperate environments, distributed between 33°34' and 38°41' S. The second experiment, aiming to identify potential physiological traits linked to the different yields of the two crops, was conducted in two Mediterranean sites (Cauquenes and Santa Rosa) in which crops were grown under well-watered (WW) and water-limited (WL) conditions. The relationship between GY and the environmental index revealed that triticale exhibited a higher regression coefficient (Finlay and Wilkinson slope), indicating a more stable response to the environment, accompanied by higher yields than bread wheat. Harvest index was not significantly different between the two cereals, but triticale had higher kernels per spike (35%) and 1000 kernel weight (16%) than wheat, despite a lower number of spikes per square meter. The higher yield of triticale was linked to higher values of chlorophyll content, leaf net photosynthesis (An), the maximum rate of electron transport (ETRmax), the photochemical quantum yield of PSII [Y(II)] and leaf water-use efficiency. GY was positively correlated with Ci at anthesis and Δ13C in both species, as well as with gs at anthesis in triticale, but negatively correlated with non-photochemical fluorescence quenching and quantum yield of non-photochemical energy conversion at grain filling in wheat. These results revealed that triticale presented higher photosynthetic rates that contributed to increase plant growth and yield in the different environments, whereas wheat showed higher photoprotection system in detriment of assimilate production.

Assessing Wheat Traits by Spectral Reflectance: Do We Really Need to Focus on Predicted Trait-Values or Directly Identify the Elite Genotypes Group?

Phenotyping, via remote and proximal sensing techniques, of the agronomic and physiological traits associated with yield potential and drought adaptation could contribute to improvements in breeding programs. In the present study, 384 genotypes of wheat (Triticum aestivum L.) were tested under fully irrigated (FI) and water stress (WS) conditions. The following traits were evaluated and assessed via spectral reflectance: Grain yield (GY), spikes per square meter (SM2), kernels per spike (KPS), thousand-kernel weight (TKW), chlorophyll content (SPAD), stem water soluble carbohydrate concentration and content (WSC and WSCC, respectively), carbon isotope discrimination (Δ13C), and leaf area index (LAI). The performances of spectral reflectance indices (SRIs), four regression algorithms (PCR, PLSR, ridge regression RR, and SVR), and three classification methods (PCA-LDA, PLS-DA, and kNN) were evaluated for the prediction of each trait. For the classification approaches, two classes were established for each trait: The lower 80% of the trait variability range (Class 1) and the remaining 20% (Class 2 or elite genotypes). Both the SRIs and regression methods performed better when data from FI and WS were combined. The traits that were best estimated by SRIs and regression methods were GY and Δ13C. For most traits and conditions, the estimations provided by RR and SVR were the same, or better than, those provided by the SRIs. PLS-DA showed the best performance among the categorical methods and, unlike the SRI and regression models, most traits were relatively well-classified within a specific hydric condition (FI or WS), proving that classification approach is an effective tool to be explored in future studies related to genotype selection.

Cloning and functional characterization of HKT1 and AKT1 genes of Fragaria spp.-Relationship to plant response to salt stress.

Commercial strawberry, Fragaria x ananassa Duch., is a species sensitive to salinity. Under saline conditions, Na+ uptake by the plant is increased, while K+ uptake is significantly reduced. Maintaining an adequate K+/Na+ cytosolic ratio determines the ability of the plant to survive in saline environments. The goal of the present work was to clone and functionally characterize the genes AKT1 and HKT1 involved in K+ and Na+ transport in strawberry and to determine the relationship of these genes with the responses of three Fragaria spp. genotypes having different ecological adaptations to salt stress. FaHKT1 and FcHKT1 proteins from F. x ananassa and F. chiloensis have 98.1% of identity, while FaAKT1 and FcAKT1 identity is 99.7%. FaHKT1 and FaAKT1 from F. x ananassa, were functionally characterized in Xenopus oocytes. FaHKT1, belongs to the group I of HKT transporters and is selective for Na+. Expression of FaAKT1 in oocytes showed that the protein is a typical inward-rectifying and highly K+-selective channel. The relative expression of Fragaria HKT1 and AKT1 genes was studied in roots of F. x ananassa cv. Camarosa and of F. chiloensis (accessions Bau and Cucao) grown under salt stress. The expression of AKT1 was transiently increased in 'Camarosa', decreased in 'Cucao' and was not affected in 'Bau' upon salt stress. HKT1 expression was significantly increased in roots of 'Cucao' and was not affected in the other two genotypes. The increased relative expression of HKT1 and decreased expression of AKT1 in 'Cucao' roots correlates with the higher tolerance to salinity of this genotype in comparison with 'Camarosa' and 'Bau'.

Chlorophyll, anthocyanin, and gas exchange changes assessed by spectroradiometry in Fragaria chiloensis under salt stress.

Chlorophyll and anthocyanin contents provide a valuable indicator of the status of a plant's physiology, but to be more widely utilized it needs to be assessed easily and non-destructively. This is particularly evident in terms of assessing and exploiting germplasm for plant-breeding programs. We report, for the first time, experiments with Fragaria chiloensis (L.) Duch. and the estimation of the effects of response to salinity stress (0, 30, and 60 mmol NaCl/L) in terms of these pigments content and gas exchange. It is shown that both pigments (which interestingly, themselves show a high correlation) give a good indication of stress response. Both pigments can be accurately predicted using spectral reflectance indices (SRI); however, the accuracy of the predictions was slightly improved using multilinear regression analysis models and genetic algorithm analysis. Specifically for chlorophyll content, unlike other species, the use of published SRI gave better indications of stress response than Normalized Difference Vegetation Index. The effect of salt on gas exchange is only evident at the highest concentration and some SRI gave better prediction performance than the known Photochemical Reflectance Index. This information will therefore be useful for identifying tolerant genotypes to salt stress for incorporation in breeding programs.

Insights into the structure of urea-like compounds as inhibitors of the juvenile hormone epoxide hydrolase (JHEH) of the tobacco hornworm Manduca sexta: analysis of the binding modes and structure-activity relationships of the inhibitors by docking and CoMFA calculations.

Substituted urea compounds are well-known as potent inhibitors of juvenile hormone epoxide hydrolase (JHEH) of the tobacco hornworm Manduca sexta. Docking simulations of 47 derivatives inside JHEH were performed to gain insight into the structural characteristics of these complexes. The obtained orientations show a strong similitude with the observed in the known X-ray crystal structures of human soluble epoxide hydrolase (sEH) complexed with dialkylurea inhibitors. In addition, the predicted inhibitor concentration (IC50) of the above-mentioned compounds as JHEH inhibitors were obtained by a quantitative structure-activity relationship (QSAR) method by using comparative molecular field analysis (CoMFA) applied to aligned dataset. The best models included steric and electrostatic fields and had adequate predictive abilities. In addition, these models were used to predict the activity of an external test set of compounds that was not used for building the model. Furthermore, plots of the CoMFA fields allowed conclusions to be drawn for the choice of suitable inhibitors.

Classification of voltage-gated K(+) ion channels from 3D pseudo-folding graph representation of protein sequences using genetic algorithm-optimized support vector machines.

Voltage-gated K(+) ion channels (VKCs) are membrane proteins that regulate the passage of potassium ions through membranes. This work reports a classification scheme of VKCs according to the signs of three electrophysiological variables: activation threshold voltage (V(t)), half-activation voltage (V(a50)) and half-inactivation voltage (V(h50)). A novel 3D pseudo-folding graph representation of protein sequences encoded the VKC sequences. Amino acid pseudo-folding 3D distances count (AAp3DC) descriptors, calculated from the Euclidean distances matrices (EDMs) were tested for building the classifiers. Genetic algorithm (GA)-optimized support vector machines (SVMs) with a radial basis function (RBF) kernel well discriminated between VKCs having negative and positive/zero V(t), V(a50) and V(h50) values with overall accuracies about 80, 90 and 86%, respectively, in crossvalidation test. We found contributions of the "pseudo-core" and "pseudo-surface" of the 3D pseudo-folded proteins to the discrimination between VKCs according to the three electrophysiological variables.

Protein radial distribution function (P-RDF) and Bayesian-Regularized Genetic Neural Networks for modeling protein conformational stability: chymotrypsin inhibitor 2 mutants.

Development of novel computational approaches for modeling protein properties is a main goal in applied Proteomics. In this work, we reported the extension of the radial distribution function (RDF) scores formalism to proteins for encoding 3D structural information with modeling purposes. Protein-RDF (P-RDF) scores measure spherical distributions on protein 3D structure of 48 amino acids/residues properties selected from the AAindex data base. P-RDF scores were tested for building predictive models of the change of thermal unfolding Gibbs free energy change (DeltaDeltaG) of chymotrypsin inhibitor 2 upon mutations. In this sense, an ensemble of Bayesian-Regularized Genetic Neural Networks (BRGNNs) yielded an optimum nonlinear model for the conformational stability. The ensemble predictor described about 84% and 70% variance of the data in training and test sets, respectively.

Proteometric study of ghrelin receptor function variations upon mutations using amino acid sequence autocorrelation vectors and genetic algorithm-based least square support vector machines.

Functional variations on the human ghrelin receptor upon mutations have been associated with a syndrome of short stature and obesity, of which the obesity appears to develop around puberty. In this work, we reported a proteometrics analysis of the constitutive and ghrelin-induced activities of wild-type and mutant ghrelin receptors using amino acid sequence autocorrelation (AASA) approach for protein structural information encoding. AASA vectors were calculated by measuring the autocorrelations at sequence lags ranging from 1 to 15 on the protein primary structure of 48 amino acid/residue properties selected from the AAindex database. Genetic algorithm-based multilinear regression analysis (GA-MRA) and genetic algorithm-based least square support vector machines (GA-LSSVM) were used for building linear and non-linear models of the receptor activity. A genetic optimized radial basis function (RBF) kernel yielded the optimum GA-LSSVM models describing 88% and 95% of the cross-validation variance for the constitutive and ghrelin-induced activities, respectively. AASA vectors in the optimum models mainly appeared weighted by hydrophobicity-related properties. However, differently to the constitutive activity, the ghrelin-induced activity was also highly dependent of the steric features of the receptor.

2D autocorrelation modelling of the inhibitory activity of cytokinin-derived cyclin-dependent kinase inhibitors.

The inhibitory activity towards p34(cdc2)/cyclin b kinase (CBK) enzyme of 30 cytokinin-derived compounds has been successfully modelled using 2D spatial autocorrelation vectors. Predictive linear and non-linear models were obtained by forward stepwise multi-linear regression analysis (MRA) and artificial neural network (ANN) approaches respectively. A variable selection routine that selected relevant non-linear information from the data set was employed prior to networks training. The best ANN with three input variables was able to explain about 87% data variance in comparison with 80% by the linear equation using the same number of descriptors. Similarly, the neural network had higher predictive power. The MRA model showed a linear dependence between the inhibitory activities and the spatial distributions of masses, electronegativities and van der Waals volumes on the inhibitors molecules. Meanwhile, ANN model evidenced the occurrence of non-linear relationships between the inhibitory activity and the mass distribution at different topological distance on the cytokinin-derived compounds. Furthermore, inhibitors were well distributed regarding its activity levels in a Kohonen self-organizing map (SOM) built using the input variables of the best neural network.

2D Autocorrelation modeling of the negative inotropic activity of calcium entry blockers using Bayesian-regularized genetic neural networks.

Negative inotropic potency of 60 benzothiazepine-like calcium entry blockers (CEBs), Diltiazem analogs, was successfully modeled using Bayesian-regularized genetic neural networks (BRGNNs) and 2D autocorrelation vectors. This approach yielded reliable and robust models whilst by means of a linear genetic algorithm (GA) search routine no multilinear regression model was found describing more than 50% of the training set. On the contrary, the optimum neural network predictor with five inputs described about 84% and 65% variances of 50 randomly selected training and test sets. Autocorrelation vectors in the nonlinear model contained information regarding 2D spatial distributions on the CEB structure of van der Waals volumes, electronegativities, and polarizabilities. However, a sensitivity analysis of the network inputs pointed out to the electronegativity and polarizability 2D topological distributions at substructural fragments of sizes 3 and 4 as the most relevant features governing the nonlinear modeling of the negative inotropic potency.

Genetic neural network modeling of the selective inhibition of the intermediate-conductance Ca2+ -activated K+ channel by some triarylmethanes using topological charge indexes descriptors.

Selective inhibition of the intermediate-conductance Ca(2+)-activated K(+ )channel (IK (Ca)) by some clotrimazole analogs has been successfully modeled using topological charge indexes (TCI) and genetic neural networks (GNNs). A neural network monitoring scheme evidenced a highly non-linear dependence between the IK (Ca) blocking activity and TCI descriptors. Suitable subsets of descriptors were selected by means of genetic algorithm. Bayesian regularization was implemented in the network training function with the aim of assuring good generalization qualities to the predictors. GNNs were able to yield a reliable predictor that explained about 97% data variance with good predictive ability. On the contrary, the best multivariate linear equation with descriptors selected by linear genetic search, only explained about 60%. In spite of when using the descriptors from the linear equations to train neural networks yielded higher fitted models, such networks were very unstable and had relative low predictive ability. However, the best GNN BRANN 2 had a Q ( 2 ) of LOO of cross-validation equal to 0.901 and at the same time exhibited outstanding stability when calculating 80 randomly constructed training/test sets partitions. Our model suggested that structural fragments of size three and seven have relevant influence on the inhibitory potency of the studied IK (Ca) channel blockers. Furthermore, inhibitors were well distributed regarding its activity levels in a Kohonen self-organizing map (KSOM) built using the inputs of the best neural network predictor.