Seleção em progênies S1 e S2 de pessegueiro via modelos mistos (REML/BLUP)

• Autor(a) principal: Silva, José Osmar da Costa e
• Data de Publicação: 2013
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/1219
• topic: Prunus persica; Modelos mistos; Valores genéticos; Prunus persica; Mixed models; Genetic values; CNPQ::CIENCIAS AGRARIAS::AGRONOMIA::FITOTECNIA

The peach is a temperate fruit tree. It needs low winter temperatures to overcome rest and provide satisfactory production. The cultivation extends to areas of mild winters due to selection of low chilling requirement genotypes by the breeding programs. The climate adaptation and improvement of fruit quality are the main goals of peach breeding for warmer regions. This work had as general objective the estimate of genetic parameters and the prediction of breeding values of the peach breeding program at the Universidade Federal de Viçosa through the methodology of mixed models (REML/BLUP) aiming at the selection of genotypes with low chilling requirement. The specific objectives were: (i) estimate genetic parameters, to predict genetic values, estimate the genetic correlation among traits and assess the genetic variability in S1 peach populations; (ii) estimate genetic parameters, repeatability coefficient in S2 peach populations; and (iii) estimate genetic parameters, predict genotypic values, estimate the genetic correlation among traits and assess the genetic variability in peach clones. Fruit and plant traits were evaluated. The main parameters estimated were genetic variance components, the coefficients of variation and heritability. The genetic coefficients of variation were greater than 10% in most cases, considered sufficient for effective selection. In general, the heritability estimates were of moderate magnitude. The lower heritability estimate was obtained for prominence of suture region of the fruit (0.08) in S 2 populations. Some variables analyzed presented genetic correlation of high magnitude, such as plant height and trunk diameter (0.9402), number of nodes per proleptic branch and number of fruits per plant (0.7334), fruit mass with mean fruit diameter (0.9870) and fruit length (0.8550). The variable node density per meter of branch has moderately high negative correlation with plant height (-0.5613) and trunk diameter (-0.5373), what is favorable for the selection of genotypes of low vigor. The repeatability coefficients obtained for biometric characteristics of fruits in S2 populations ranged from 0.37 to 0.76. viiiThe evaluation of six fruits per plant presented a determination coefficient above 80% for the real value of permanent phenotype of the genotypes for the variables length, diameter, length/diameter ratio, percentage of red skin coverage, and prominence of the fruit apex in S2 populations. Genetic variability was observed in the different populations evaluated that enables the selection of superior genotypes for key traits. These genotypes can be used as parents for future hybridization and, or, to be released as new cultivars indicated for mild winters climates after evaluations with a larger number of repetitions in most years and locations.