Sex type determination in papaya seeds and leaves using near infrared spectroscopy combined with multivariate techniques and machine learning

Papaya trees (Carica papaya L.) can bear female, hermaphrodite, and male flowers. However, only the hermaphrodite type produces elongate fruit required and appreciated by the consumer market. Sex type determination is carried out based on the plant phenotype after planting the seedlings, increasing...

Descripción completa

Detalles Bibliográficos
Autores: Fernandes, Thiago Feliph Silva [UNESP], Silva, Raíssa Vanessa de Oliveira, Freitas, Daniel Lucas Dantas de, Sanches, Alex Guimarães [UNESP], Silva, Maryelle Barros da [UNESP], Cunha Júnior, Luis Carlos, Lima, Kássio Gomes de, Teixeira, Gustavo Henrique de Almeida [UNESP]
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2022
País:Brasil
Institución:Universidade Estadual Paulista (UNESP)
Repositorio:Repositório Institucional da UNESP
Idioma:inglés
OAI Identifier:oai:repositorio.unesp.br:11449/223178
Acceso en línea:http://dx.doi.org/10.1016/j.compag.2021.106674
http://hdl.handle.net/11449/223178
Access Level:acceso abierto
Palabra clave:Carica papaya L
Chemometrics
PCA-LDA
PCA-QDA
Sexing
Descripción
Sumario:Papaya trees (Carica papaya L.) can bear female, hermaphrodite, and male flowers. However, only the hermaphrodite type produces elongate fruit required and appreciated by the consumer market. Sex type determination is carried out based on the plant phenotype after planting the seedlings, increasing the production costs. In this regard, the objective of this study was to verify the feasibility of using near infrared spectroscopy (NIR) as a non-destructive method for sexing papaya trees. The NIR spectra were collected using seeds and leaves of the respective seedlings of the cultivars ‘T2′, ‘Formosa’, and ‘Calimosa’ (Formosa group), and ‘THB’ and ‘Ouro’ (Solo group). By using the seeds, it was possible to obtain a F-score value of 0.81 for the external validation set applying the principal component analysis and quadratic discriminant analysis (PCA-QDA). By using the leaves, the F-score values was slightly lower (0.79) applying PCA and linear discriminant analysis (PCA-LDA). It was possible to use NIR spectroscopy associated with multivariate techniques as a non-destructive method to determine the sex types in papaya trees using both seeds and leaves of the seedlings.