Banca de DEFESA: THAÍSA DE OLIVEIRA SILVEIRA

Uma banca de DEFESA de MESTRADO foi cadastrada pelo programa.
STUDENT : THAÍSA DE OLIVEIRA SILVEIRA
DATE: 30/09/2021
TIME: 13:30
LOCAL: UFRRJ (sala virtual)
TITLE:

Genetic diversity among bitter gourd (Momordica charantia L.) genotypes accessed by morphoagronomic variables

KEY WORDS:

Plant breeding; medicinal plant; germplasms; PCA, K-means grouping; phenotypic correlation; Diversity index.

PAGES: 66
BIG AREA: Ciências Agrárias
AREA: Agronomia
SUBÁREA: Fitotecnia
SPECIALTY: Melhoramento Vegetal
SUMMARY:

The bitter gourd (Momordica charantia L.; Cucurbiateceae) is an important medicinal species, recommended by ANVISA as a hypoglycemic agent. The characterization of accessions in germplasm collections is a crucial step in any breeding program, especially for those still considered non-domesticated. This is the case of the bitter gourd. The present work aimed to propose a list of descriptors based on morphoagronomic characteristics for the bitter gourd, and use it to know the genetic diversity of the germplasm collection of the species from DFITO/IA/UFRRJ. The ultimate aim is to identify genotypes to be incorporated into species improvement programs. To this end, a list of 60 descriptors, unpublished and transcribed from the literature, was designed and standardized, covering all parts of the plant: branches, leaves, flowers, fruits, seeds and arils. Then, the characterization of 88 plants from the collection was carried out, all arranged in pots in the greenhouse, in Seropédica, RJ. All quantitative variables were subjected to descriptive analysis. Then, redundant descriptors were discarded via PCA. Descriptors with greater relative contribution in components 1 and 2 were maintained. In possession of a new matrix, the dispersion analysis was carried out via PCA. In the two-dimensional plan, we sought to understand the behavior of the genotypes (UFRRJ MSC's), individually, in relation to the remaining descriptors. From the K-means statistic, patterns or groups of plants were defined within the two-dimensional plane in the PCA. In these groups, the plants were analyzed in relation to their respective origins (by State and by brazilian region). In addition, the groups were studied as to their standard phenotype, using radar-type graphics. The intra-group k-means diversity was verified via similarity analysis (ANOSIM). The relative importance of the variables was estimated using the Singh method, and the population genetic diversity (total) was verified using the Shannon and Pielou indices. 39 quantitative descriptors were created, 11 binary and 10 multicategorical. The average weight of fresh fruits and the number of female flowers stood out for their wide variation. In all, 18 variables were redundant. The PCA showed that plants with smaller fruits were the most productive. The most productive genotypes were located close to the two-dimensional plane. The genotypes UFRRJ MSC072, 042, 028 and 087 stood out for the number of fruits produced. Five statistically distinct groups were defined using the k-mens algorithm. Groups G1 and G5 were antagonistic regarding the production of fruits and seeds and regarding the sizes of fruits, leaves and seeds. The radar-type graphics allowed a panoramic view of the behavior of variables within each group. A tendency of reduction in the size of fruits, leaves and seeds was identified, occurring from group G1 to G5. The similarity analysis (ANOSIM) showed that there were statistically significant differences between the groups. The variable number of male flowers (NMf) was identified as the one with the greatest contribution to the estimation of diversity in the plant collection. Genetic diversity was considered high. Genotypes with desirable traits were identified and crosses between their progenies are indicated in the bitter gourd breeding program at UFRRJ.

BANKING MEMBERS:
Presidente - 1712769 - PEDRO CORREA DAMASCENO JUNIOR
Interna - 2271445 - BRUNA RAFAELA DA SILVA MENEZES
Externa à Instituição - MARIA LUIZA DE ARAÚJO - PESAGRO-RIO