Early detection of resistance to Mycosphaerella fijiensis in local genotypes of Musa in Ecuador

Abstract

ResumenEl objetivo de la investigación fue determinar la reacción de 14 genotipos de musáceas frente a Mycosphaerella fijensis, agente causal de la Sigatoka negra. Se condujeron dos experimentos en el Departamento de Protección Vegetal de la Estación Experimental Pichilingue del Instituto Nacional de Investigaciones Agropecuarias (INIAP), Ecuador. Los genotipos evaluados fueron: banano Orito (AA), Guineo de jardín, Williams, Filipino, Gros Michel, Mulato y Morado, del genoma (AAA), y los plátanos Limeño, Maqueño, Dominico, Dominico-Hartón, Dominico gigante, Dominico negro y Barraganete, del genoma AAB. Se usaron dos métodos de inoculación: a) in vitro sobre fragmentos de hoja con una suspensión de conidios y b) inoculación de plántulas de ocho semanas de edad con una suspensión micelial. En el primer caso se midió el periodo de incubación (PI) y el número de lesiones a los 14 y 20 días después de la inoculación (DDI). En invernadero, se registró el PI, tiempo de evolución de síntomas (TES) y tiempo de desarrollo de la enfermedad (TDE). Se determinaron diferencias estadísticas significativas (p ≤ 0,05) para todas las variables evaluadas; los cultivares Orito (AA) y Limeño (AAB) calificaron como resistentes, mientras que todos los demás fueron susceptibles, aunque mostraron un rango de variabilidad en la respuesta a las inoculaciones. Palabras clave: cultivares de Musa; inoculación; inóculo; resistencia; biodiversidad. AbstractThe aim of this research was to determine resistance reaction of 14 Musa genotypes against Mycosphaerella fijensis, causal agent of Black Sigatoka disease. Two experiments were conducted in laboratory and shadehause at Pichilingue research Station from the National Institute of Agricultural Research (INIAP), Ecuador. Genotypes considered were: Orito (AA), Guineo de jardín, Williams, Filipino, Gross Mitchel, Mulato and Morado, (AAA); Limeño, Maqueño, Dominico, Dominico-Hartón, Dominico gigante, Dominico negro and Barraganete from the AAB genome. Two inoculation methods were used to asses' resistance: a) in vitro inoculation over leaf fragments with a conidial suspension, and b) inoculation of eight weeks old seedlings with a mycelial suspension of the fungus. In the first case, the incubation period (IP) and the number of lesions at 14 and 20 days after inoculation (DDI) were measured. For in vitro trial, the incubation period (IP) and the number of leaf lesions 14 and 20 days after inoculation (DDI) were measured, and for seedlings, PI, period of symptoms evolution (TES) and time for disease development (TDE) were recorded. Statistically significant (p ≤ 0.05) differences were found for all variables. Cultivars Orito (AA) and Limeño (AAB) qualify as resistant while all the others were susceptible, though response to inoculation within the latter group showed a range of variability.

Potential Uses of Musaceae Wastes: Case of Application in the Development of Bio-Based Composites

“…In the case of bananas, the primary producers are (1) India, (2) China, (3) Indonesia, (4) Brazil, (5) Ecuador, (6) the Philippines, (7) Angola, (8) Guatemala, and (9) Colombia [ 18 ]. In Colombia, the plantain species of greater use for cultivation are Dominico-Hartón, Dominico, and Hartón from the use of 405,910 hectares in the planted area, generating a production of 3,534,083 tons and a production yield of 8.71 tons/hectare [ 15 , 16 ], while the Banana, Cavendish, and Gros Michel species are harvested from the use of 82,905 hectares, producing 2,052,991 tons and a yield of 24.75 tons/hectare in the year 2015 [ 15 , 16 , 19 , 20 ].…”

“… Type of Musaceae; 1 Sweet banana; 2 Banana; 3 Cooking banana [ 7 , 14 , 19 , 21 , 22 , 23 , 24 , 25 ]. …”

Black Sigatoka Classification Using Convolutional Neural Networks

“…A further study introduced the application of pre-training networks such as ResNet50 and Inception-v3, which generated efficiencies of 99.9% [14]. Even though those efficacies have high accuracy, these methodologies only detect the disease while skipping metrics that define different levels of severalty like the Fouré scale [15] and each stage must carry a different action control. Disease detection approaches using CNNs classify diseases using images with a resolution of 64 × 64 to 150 × 150 [16], whereas current mobile devices have cameras with average resolutions of 4160 × 3120.…”

Black Sigatoka Classification Using Convolutional Neural Networks

“…A further study introduced the application of pre-training networks such as ResNet50 and Inception-v3, which generated efficiencies of 99.9% [11]. Even though those efficacies have high accuracy, these methodologies only detect the disease while skipping metrics that define different levels of severalty like the Fouré scale [12] and each stage must carry a different action control. Disease detection approaches using CNNs classify diseases using images with a resolution of 64 × 64 to 150 × 150 [13], whereas current mobile devices have cameras with average resolutions of 4160 × 3120.…”