Machine learning applied to the prediction of root architecture of soybean cultivars under two water availability conditions / Machine learning aplicado à predição da arquitetura radicular de cultivares de soja sob duas condições de disponibilidade hídrica

The objective of this study was to evaluate the performance of four machine learning models, as well as multitask learning, to predict soybean root variables from simpler variables, under two water availability conditions. In order to do so, 100 soybean cultivars were conducted in a greenhouse under a control condition and a stress condition. Aerial part and root variables were evaluated. The machine learning models used to predict complex root variables were artificial neural network (ANN), random forest (RF), extreme gradient boosting (EGBoost) and support vector machine (SVM). A linear model was used for comparison purposes. Multitask learning was employed for ANN and RF. In addition, feature importance was defined using RF and XGBoost algorithms. All the machine learning models performed better than the linear model. In general, SVM had the greatest potential for the prediction of most of the root variables, with better values of RMSE, MAE and R2. Dry weight of the aerial part and root volume exhibited the greatest importance in the predictions. The models developed using multitask learning performed similarly to the ones conventionally developed. Finally, it is concluded that the machine learning models evaluated can be used to predict root variables of soybean from easily measurable variables, such as dry weight of the aerial part and root volume.(AU)

O objetivo deste estudo foi avaliar o desempenho de quatro modelos de machine learning, bem como multitask learning, para predizer variáveis radiculares de soja a partir de variáveis simples, em duas condições de disponibilidade hídrica. Para isso,100 cultivares de soja foram conduzidas em casa de vegetação sob uma condição controle e uma condição estresse. Foram avaliadas as variáveis da parte aérea e da raiz. Os modelos machine learning usados para predizer variáveis complexas do sistema radicular foram rede neural artificial (RNA), random forest (RF), extreme gradient boosting (EGBoost) e support vector machine (SVM). O modelo linear foi usado para fins de comparação. O multitask learning foi empregado para RNA e RF. Além disso, a importância das variáveis foi definida usando algoritmos RF e XGBoost. Todos os modelos de machine learning apresentaram melhor desempenho do que o modelo linear. Em geral, SVM apresentou o maior potencial de predição da maioria das variáveis raiz, com melhores valores de RMSE, MAE e R2. O peso seco da parte aérea e o volume da raiz exibiram as maiores importâncias nas predições. Os modelos desenvolvidos por meio do multitask learning apresentaram desempenhos semelhantes aos desenvolvidos convencionalmente. Por fim, conclui-se que os modelos de machine learning avaliados podem ser usados para predizer variáveis radiculares de soja a partir de variáveis facilmente mensuráveis, como massa seca da parte aérea e volume radicular.(AU)

Weathering deterioration in pre-harvest of soybean seeds: physiological, physical, and morpho-anatomical changes

ABSTRACT: Successive cycles of water absorption and loss favor weathering deterioration, one of the main factors that affect the quality of soybean seeds. This study evaluated the physiological, physical, and morpho-anatomical changes in soybean seeds under weathering deterioration at the pre-harvest phase. Six soybean cultivars (BMX Apolo, DM 6563, NS 5959, NA 5909, BMX Potência, and TMG 1175) were produced in a greenhouse and underwent weathering deterioration through a rainfall simulation system, applying 0, 60, 120, and 180 mm of precipitation at pre-harvest phase. Each rainfall level was divided into two applications at an interval of 72 h: 60 mm (30 + 30), 120 mm (60 + 60), and 180 mm (90 + 90). After harvest, the seeds were evaluated for germination, vigor, physical and morpho-anatomical properties. Weathering deterioration induced by simulated rainfall at the pre-harvest phase contributes to the reduction in soybean seed germination and vigor and is conditioned by the soybean genotype. The increase in intensity of simulated rainfall led to a more significant weathering damage in seeds, as evidenced by the X-ray and tetrazolium test. Cultivars DM 6563 and BMX Potência were more susceptible, while NA 5909 was less susceptible to weathering deterioration (especially at the highest level; 120 mm and 180 mm). Anatomical changes caused by weathering deterioration lead to cell compaction and rupture, mainly in the cell layers of the hourglass and parenchyma, forming intracellular spaces. The presence of weathering damage caused a reduction in physiological soybean seed quality.

Weathering deterioration in pre-harvest of soybean seeds: physiological, physical, and morpho-anatomical changes

Successive cycles of water absorption and loss favor weathering deterioration, one of the main factors that affect the quality of soybean seeds. This study evaluated the physiological, physical, and morpho-anatomical changes in soybean seeds under weathering deterioration at the pre-harvest phase. Six soybean cultivars (BMX Apolo, DM 6563, NS 5959, NA 5909, BMX Potência, and TMG 1175) were produced in a greenhouse and underwent weathering deterioration through a rainfall simulation system, applying 0, 60, 120, and 180 mm of precipitation at pre-harvest phase. Each rainfall level was divided into two applications at an interval of 72 h: 60 mm (30 + 30), 120 mm (60 + 60), and 180 mm (90 + 90). After harvest, the seeds were evaluated for germination, vigor, physical and morpho-anatomical properties. Weathering deterioration induced by simulated rainfall at the pre-harvest phase contributes to the reduction in soybean seed germination and vigor and is conditioned by the soybean genotype. The increase in intensity of simulated rainfall led to a more significant weathering damage in seeds, as evidenced by the X-ray and tetrazolium test. Cultivars DM 6563 and BMX Potência were more susceptible, while NA 5909 was less susceptible to weathering deterioration (especially at the highest level; 120 mm and 180 mm). Anatomical changes caused by weathering deterioration lead to cell compaction and rupture, mainly in the cell layers of the hourglass and parenchyma, forming intracellular spaces. The presence of weathering damage caused a reduction in physiological soybean seed quality.

Weathering deterioration in pre-harvest of soybean seeds: physiological, physical, and morpho-anatomical changes

Successive cycles of water absorption and loss favor weathering deterioration, one of the main factors that affect the quality of soybean seeds. This study evaluated the physiological, physical, and morpho-anatomical changes in soybean seeds under weathering deterioration at the pre-harvest phase. Six soybean cultivars (BMX Apolo, DM 6563, NS 5959, NA 5909, BMX Potência, and TMG 1175) were produced in a greenhouse and underwent weathering deterioration through a rainfall simulation system, applying 0, 60, 120, and 180 mm of precipitation at pre-harvest phase. Each rainfall level was divided into two applications at an interval of 72 h: 60 mm (30 + 30), 120 mm (60 + 60), and 180 mm (90 + 90). After harvest, the seeds were evaluated for germination, vigor, physical and morpho-anatomical properties. Weathering deterioration induced by simulated rainfall at the pre-harvest phase contributes to the reduction in soybean seed germination and vigor and is conditioned by the soybean genotype. The increase in intensity of simulated rainfall led to a more significant weathering damage in seeds, as evidenced by the X-ray and tetrazolium test. Cultivars DM 6563 and BMX Potência were more susceptible, while NA 5909 was less susceptible to weathering deterioration (especially at the highest level; 120 mm and 180 mm). Anatomical changes caused by weathering deterioration lead to cell compaction and rupture, mainly in the cell layers of the hourglass and parenchyma, forming intracellular spaces. The presence of weathering damage caused a reduction in physiological soybean seed quality.(AU)

Soybean productivity, stability, and adaptability through mixed model methodology / Produtividade, estabilidade e adaptabilidade da soja por meio de metodologia de modelo misto

The genotype × environment (G×E) interaction plays an essential role in phenotypic expression and can lead to difficulties in genetic selection. Thus, the present study aimed to estimate genetic parameters and to compare different selection strategies in the context of mixed models for soybean breeding. For this, data referring to the evaluation of 30 genotypes in 10 environments, regarding the grain yield trait, were used. The variance components were estimated through restricted maximum likelihood (REML) and genotypic values were predicted through best linear unbiased prediction (BLUP). Significant effects of genotypes and G×E interaction were detected by the likelihood ratio test (LRT). Low genotypic correlation was obtained across environments, indicating complex G×E interaction. The selective accuracy was very high, indicating high reliability. Our results showed that the most productive soybean genotypes have high adaptability and stability.(AU)

A interação genótipo × ambiente (G × E) desempenha um papel essencial na expressão fenotípica e pode provocar dificuldades na seleção genética. Assim, o presente estudo teve como objetivo estimar parâmetros genéticos e comparar diferentes estratégias de seleção no contexto de modelos mistos para melhoramento da soja. Para isso, foram utilizados dados referentes à avaliação de 30 genótipos em dez ambientes, referentes à característica produtividade de grãos. Os componentes de variância foram estimados pela máxima verossimilhança restrita (REML) e os valores genotípicos foram preditos pela melhor previsão imparcial linear (BLUP). Efeitos significativos dos genótipos e interação G × E foram detectados pelo teste da razão de verossimilhança (LRT). Correlação genotípica baixa foi obtida entre os ambientes indicando interação G × E do tipo complexa. A acurácia seletiva foi muito alta, indicando alta confiabilidade. Os resultados mostraram que os genótipos de soja mais produtivos apresentam alta adaptabilidade e estabilidade.(AU)

Seleção assistida por marcadores para teor de matéria seca e açúcares redutores em tubérculos de batata / Marker assisted selection for dry matter and reducing sugars in potato tubers

A seleção precoce de clones que possuam níveis elevados de matéria seca e baixos teores de açúcares redutores é uma necessidade nos programas de melhoramento para a qualidade de processamento da batata (Solanum tuberosum L.) na forma de palitos fritos ou chips. A seleção precoce tornou-se possível com a utilização de marcadores genéticos, visto que permitem a identificação precisa de indivíduos superiores. Assim, procura-se cada vez mais encontrar marcadores capazes de caracterizar tais indivíduos e utilizá-los via seleção assistida. O objetivo deste trabalho foi avaliar a eficiência da seleção assistida, utilizando os marcadores identificados por ANDREU (2004) que estariam associados ao teor de matéria seca e açúcares redutores em tubérculos de batata. Clones provenientes de 20 famílias foram avaliados nas gerações de plântula (P), primeira geração clonal (C1) e segunda geração clonal (C2). As estimativas das correlações simples para os caracteres entre gerações foram significativas, porém, baixas, confirmando a inviabilidade de se efetuar a seleção precoce nas primeiras gerações com base apenas em informações fenotípicas. Os marcadores utilizados forneceram um total de 16 marcas. Pela regressão múltipla stepwise, apenas sete dessas marcas tiveram associação com os caracteres estudados. Além disso, nenhuma marca associada ao teor de matéria seca de tubérculos na geração C1 teve associação significativa na geração C2. Isso também foi observado com o teor de açúcares redutores, o que é um indicativo da interação QTLs x ambientes. A seleção assistida não se mostrou eficiente em relação à fenotípica em nenhum dos casos avaliados, portanto, não sendo útil em uma possível seleção precoce. Esses resultados indicam que tais marcadores não estão próximos aos genes controladores dos caracteres desejados, sendo necessária a identificação de novos marcadores mais associados que possibilitem maior eficiência da seleção assistida.(AU)

Early generation selection for clones with high content of tuber dry matter and low levels of reducing sugars is required for potato (Solanum tuberosum L.) processing. Selection of superior clones at early generations became possible with the deployment of genetic markers, and can precisely identify the superior individuals. Therefore, it is necessary to identify genetic markers closely linked to genes of interest to do assisted selection. The aim of this research was to evaluate the efficiency of marker assisted selection with genetic markers previously identified by ANDREU (2004), which are assumed to be associated with dry matter and reducing sugars content in potato tubers. Clones from 20 families were evaluated during the seedling generation (S), first clonal generation (C1) and second clonal generation (C2). The estimated coefficients of correlation for all traits among generations were significant, even though of low magnitude, confirming that selection at early generation based only on phenotypic traits is inviable. A total of sixteen bands were amplified using these markers. However, by multiple stepwise regression, only seven of these bands showed association with the evaluated traits. Moreover, no markers associated with dry matter and reducing sugars content in the C1 were significantly associated with these traits in the C2, suggesting the existence of QTLs x environment interactions. The marker assisted selection resulted less efficient than the phenotypic selection in all cases studied, and thus is not recommended for early generation selection of clones for the processing industry. These results suggest that the markers used are not closely linked to the genes controlling the traits most important for processing. Therefore, it is important to identify new markers closely linked with such traits of interest that could improve the efficiency of marker assisted selection.(AU)