Felipe de Oliveira Dias

Felipe de Oliveira Dias

Auto-generated profile, based on publicly available information

Felipe de Oliveira Dias

Affiliation Department of Agronomy, Universidade Federal de Viçosa, Viçosa, Minas Gerais, Brazil

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 March 2023 in International Journal of Remote Sensing
Verified
Journal Article
Closed access
FDFelipe de Oliveira Dias
Felipe de Oliveira Dias
Domingos Sarvio Magalhães ValenteDomingos Sarvio Magalhães Valente
Domingos Sarvio Magalhães Valente
COCarolina Tavares Oliveira
Carolina Tavares Oliveira
FDFrançoise Dalprá Dariva
Françoise Dalprá Dariva
MCMariane Gonçalves Ferreira Copati
Mariane Gonçalves Ferreira Copati
CNCarlos Nick
Carlos Nick
The selection of late blight resistant genotypes of tomato requires many evaluations on the field. The use of machine learning models to assess late blight severity based on images from multispectral cameras onboard of unmanned aerial vehicles (UAVs) can bring efficiency and quickly in evaluations. In this work, we use remote sensing and machine learning techniques to assess late blight resistance of tomato lines grown in open field conditions. Seventy-six tomato lines, including two resistant lines and one susceptible line, were used to quantify late blight severity. Plants were arranged according to a replicated check design in a total of 132 experimental plots. Tomato plants were artificially inoculated with a Phytophthora infestans zoospore suspension. Multispectral images were obtained using an unmanned aerial vehicle. We calculated vegetation indexes (VI) using the images, which were the basis for building the Random Forest models used to predict disease severity. Two methodologies were used to predict late blight severity: Methodology 1, which used only the images from the last day of evaluation, and Methodology 2, which used the images from four days of evaluation. For Methodology 1 and 2, determination coefficients of 0.81 and 0.93 were obtained for the test set, respectively. Methodology 2 was used to predict late blight severity of the 132 field plots. Tomato plots were sorted from lowest to highest predicted severity. Resistant plots were ranked first indicating consistency of prediction. We therefore recommend Methodology 2 as a fast and practical way to predicted late blight severity in breeding populations.

15 June 2023 in PLOS ONE
Verified
Journal Article
Open access
PDF file
HPHerika Paula Pessoa
Herika Paula Pessoa
FDFrançoise Dalprá Dariva
Françoise Dalprá Dariva
MCMariane Gonçalves Ferreira Copati
Mariane Gonçalves Ferreira Copati
RPRamon Gonçalves de Paula
Ramon Gonçalves de Paula
FDFelipe de Oliveira Dias
Felipe de Oliveira Dias
CNCarlos Nick
Carlos Nick
Tomato plants are sensitive to drought stress throughout their growth cycle. To be considered drought-tolerant, a cultivar should display tolerance at all developmental stages. This study aimed to evaluate whetherSolanum pennelliiintrogression lines (ILs) previously selected as drought-tolerant during germination/seedling growth maintained this tolerance in the vegetative/reproductive stage. We then investigated these ILs to uncover candidate genes. The plants were subjected to two different environmental conditions: well-watered and drought-stressed (water withheld for ≤ 20 d after flowering). Phenotyping for morphological, physiological, fruit quality, and yield-related traits was performed, and the data was analyzed using a mixed-model approach. Using a multi-trait index that relies on factor analysis and genotype-ideotype distance (FAI-BLUP index), the genotypes were ordered based on how far they were from the drought-tolerant ideotype. Afterward, the tomato IL population map furnished by the SOL Genomics Network was utilized to identify introgressed segments of significance for the identification of candidate genes. Significant genotypic differences were found in the yield, water content, mean weight, length, and width of the fruit, the percentage of fruits displaying blossom-end rot, and titratable acidity. The drought-tolerance ideotype was built considering the maximum values for the fruit water content, number of fruits, mean fruit weight, and yield, minimum values for blossom-end rot, and mean values for titratable acidity. IL 1-4-18, IL 7-4-1, IL 7–1, IL 7-5-5, and IL 1–2 were ranked above M-82 and therefore considered drought-tolerant during the vegetative/reproductive stage. IL 1-4-18 and IL1-2 sustained drought tolerance displayed during germination/seedling growth into the vegetative/reproductive stage. The following candidate genes associated with drought tolerance were identified:AHG2,At1g55840,PRXIIF,SAP5,REF4-RELATED 1,PRXQ,CFS1,LCD,CCD1, andSCS. Because they are already associated with genetic markers, they can be transferred to elite tomato cultivars through marker-assisted technology after validation.

1 June 2021 in Scientia Horticulturae
Verified
Journal Article
Closed access
HPHerika Paula Pessoa
Herika Paula Pessoa
LSLaércio Junio Da Silva
Laércio Junio Da Silva
RCRafael Ravanelli Chagas
Rafael Ravanelli Chagas
MCMariane Gonçalves Ferreira Copati
Mariane Gonçalves Ferreira Copati
FDFrançoise Dalprá Dariva
Françoise Dalprá Dariva
CNCarlos Nick
Carlos Nick

1 March 2019 in Horticultura Brasileira
Verified
Journal Article
Open access
PDF file
MFMariane G Ferreira
Mariane G Ferreira
GAGabriella Q de Almeida
Gabriella Q de Almeida
HPHerika P Pessoa
Herika P Pessoa
FDFrançoise D Dariva
Françoise D Dariva
FDFelipe de O Dias
Felipe de O Dias
CNCarlos Nick
Carlos Nick
This study was performed to introduce the allele Bush of commercial cultivars with a bush growth habit into Cucurbita moschata accessions and select the best crossings for production potential and standard fruit shape “Menina Brasileira” through general combining ability (GCA) and specific combining ability (SCA). To determine GCA and SCA, a partial diallel analysis was performed. The parental group (group I) consisted of two cultivars containing the dwarfism bush gene, and the second group (Group II), formed by five accessions of UFV Vegetable Germplasm Bank, with high production potential and fruit shape “Menina Brasileira”. We evaluated: productivity (PROD), fruit shape (FORM), growth rate until the 50th day after transplanting (CRESC), and length of the main stem on the 50th day (COMP). Significant differences were noticed for CRESC and COMP only between GCA of parents belonging to group I, in which Zapallo stood out. For PROD and FORM, diallel analysis proved to be efficient in studying GCA effects. Cultivar Piramoita (group I) and accessions BGH-4360 and BGH-5253 (group II) showed positive values for GCA, suggesting that these parents possess a higher frequency of favorable alleles for these two traits. Significance in SCA effects was also verified, highlighting the crossing Piramoita x BGH-4360, for PROD, indicating that this combination is the most promising for breeding purposes.

15 June 2021 in Crop Science
Verified
Journal Article
Closed access
PDF file
MCMariane Gonçalves Ferreira Copati
Mariane Gonçalves Ferreira Copati
FDFrançoise Dalprá Dariva
Françoise Dalprá Dariva
FDFelipe de Oliveira Dias
Felipe de Oliveira Dias
JRJoão Romero Do Amaral Santos de Carvalho Rocha
João Romero Do Amaral Santos de Carvalho Rocha
HPHerika Paula Pessoa
Herika Paula Pessoa
CNCarlos Nick
Carlos Nick

5 August 2025
Verified
Thesis
Open access
PDF file
FDFelipe de Oliveira Dias
Felipe de Oliveira Dias
A produção vegetal sustenta o estilo de vida das sociedades fornecendo alimentos, fibras e energia. A agricultura deve atender à crescente demanda por esses produtos, superando os desafios do clima que se impõe. O melhoramento genético vegetal é uma ciência motora da evolução da agricultura, desenvolvendo cultivares mais produtivas e adaptadas a diferentes ambientes. Muitas técnicas podem ser utilizadas na avaliação e seleção de genótipos superiores, visando alcançar objetivos diversos. O tomate é um alimento muito apreciado no mundo, e deve ser melhorado geneticamente para obter melhor performance agronômica. Um dos objetivos do melhoramento do tomateiro é obter cultivares produtivas e com qualidade do fruto. Assim, o primeiro capítulo desse trabalho visou selecionar genótipos superiores com maior produtividade e qualidade do fruto utilizando a técnica de seleção para múltiplas características através do índice de seleção FAI- BLUP. Para melhorar a acurácia seletiva, utilizou-se a técnica de modelagem da dependência espacial de experimentos de campo para selecionar o modelo de melhor ajuste para predizer os valores genotípicos para cada característica. Considerou-se três modelos de ajuste: o modelo 1 foi o modelo tradicional de análise de experimentos no delineamento de blocos casualizados, que presume independência espacial para os efeitos residuais; o modelo 2 considerou as correlações autorregressivas de primeira ordem (AR1) nas direções de linha e coluna conforme o posicionamento das parcelas experimentais no campo; e o modelo 3 que além de considerar as correlações AR1 em ambas as direções, considerou também a variação adicional independente entre parcelas (efeito nugget). Procedeu-se a seleção do modelo de melhor ajuste para cada característica pelos critérios de AIC, BIC e acurácia. Implementou-se o índice FAI-BLUP em dois cenários: modelando a dependência espacial (Cenário I) e sem a modelagem dessa dependência (Cenário II). O Cenário I foi o mais adequado para se obter ganhos genéticos conforme esperado, em direção ao ideótipo delineado. Além de obter cultivares produtivas e com qualidade do fruto, estas devem ser resistentes as principais pragas e doenças do tomateiro. Assim, o segundo capítulo desse trabalho buscou utilizar técnicas de sensoriamento remoto para avaliar genótipos de tomate em campo quanto a severidade da requeima, uma das principais doenças dessa cultura. Imagens multiespectrais do campo experimental foram obtidas com a câmera MicaSense RedEdge-MX a bordo do drone DJI, Matrice 100. As imagens foram processadas para extrair os dados das bandas espectrais e calcular os índices de vegetação. Estes índices foram utilizados para treinar modelos preditivos de aprendizado de máquinas de modo supervisionado, utilizando o algoritmo Random Forest. Duas metodologias foram consideradas: o Método 1 utilizou os índices de vegetação calculados a partir de uma única imagem obtida no último dia de avaliação; o Método 2 utilizou os índices de vegetação calculados a partir de quatro imagens obtidas em distintos dias de avaliação. Os modelos treinados com o Método 2 tiveram melhor desempenho preditivo. Os genótipos avaliados tiveram a severidade da requeima predita pelo melhor modelo selecionado após seu treinamento. As testemunhas resistentes e suscetível foram adequadamente classificadas pela severidade predita, possibilitando a seleção de genótipos de tomateiro resistentes a requeima. Palavras-chave: Solanum lycopersicum; acurácia seletiva; análise espacial; índice FAI-BLUP; imagem multiespectral; aprendizado de máquinas.