ABSTRACT The evaluation of diseases resistance is a key step in the Sugarcane Breeding Program. For this reason, it is necessary the updating of the methodologies established for this purpose. The objective of this work was to compare the current methodology and the use of linear regression for the classification of cultivars by their reaction to the sugarcane mosaic disease. Data from a multi-environmental study conducted at the three Test Centers: Jovellanos, Florida and Mayarí, were used. Thirteen sugarcane cultivars were included. The results showed variability in the magnitude of the expected response of the control, which prevented the validation of the six trials evaluated by the current methodology and, therefore, the classification of the material under study. However, the application of the comparison of means between the infection percentages of the control cultivars offered an alternative for the revalidation of the studies. In addition, the application of linear regression allowed obtaining mathematical models for the assignment of a rate and reaction category against the mosaic to the cultivars in the studies, which gives them greater precision in the results.

RESUMEN La evaluación de la resistencia a enfermedades es un paso clave en los Programas de Mejoramiento Genético de la Caña de Azúcar. Por ello se fundamenta la actualización de las metodologías establecidas con este propósito. El objetivo de este trabajo fue realizar una comparación entre la metodología vigente y el empleo de la regresión lineal para la clasificación de los cultivares por su reacción frente a la enfermedad mosaico de la caña de azúcar. Se utilizaron los datos correspondientes a un estudio multiambiental conducido en los tres Centros de Prueba: Jovellanos, Florida y Mayarí, donde se incluyeron 13 cultivares de caña de azúcar. Los resultados mostraron variabilidad en la magnitud de la respuesta esperada de los controles, lo que impidió la validación de los seis ensayos evaluados por la metodología vigente y, por tanto, la clasificación del material en estudio. Sin embargo, la aplicación de la comparación de medias entre el porcentaje de infección de los cultivares controles ofreció una alternativa para la revalidación de los estudios. Además, la aplicación de la regresión lineal permitió obtener modelos matemáticos para la asignación de grados y categorías de reacción frente al mosaico a los cultivares en estudio, lo que les confiere mayor precisión a los resultados.

ABSTRACT: Multiple pathogens affect sugarcane, among them Puccinia melanocephala, the causal agent of brown rust. This disease was first reported in Cuba in 1979 when it was responsible for a severe attack on the main sugarcane variety B4362. The aims of the present study were to give an overview of sugarcane brown rust in Cuba and show the current disease situation in the country. A retrospective analysis regard to sugarcane cultivar composition resistant to brown rust in Cuba was carried out. In addition 154 genotypes, including the most used progenitors in the breeding program and commercial varieties were evaluated under natural infection conditions. The identity of P. melanocephala was verified by Polymerase Chain Reaction (PCR) and by sequencing the ITS1 region. After the introduction of P. melanocephala into Cuba, the susceptible variety, B4362, was replaced by Ja60-5 which remained resistant until 1998. Since 2002, a varietal policy supported by a governmental resolution establishing that any single cultivar cannot occupy more than 20 % of the production area for each production company, local area and province, has been applied. Out of the genotypes evaluated, 49 showed resistance to the disease and 35 intermediate behavior, while 39 were susceptible and 31, highly susceptible. P. melanocephala was detected by PCR in all symptomatic samples and its identify confirmed by sequencing the ITS1 region. The adopted measurement together with permanent phytosanitary monitoring and commercial release of resistant or intermediate cultivars succeeded in avoiding any new epidemic. Inoculum pressure was reduced, even on susceptible and highly susceptible varieties since, by resolution, they cannot occupy more than 10 % of the planted area.