Estimação de parâmetros genéticos de características de desempenho e de rendimento de filé de grupos genéticos de tilápias do Nilo.

The objective of this study was to estimate (co ) variance, genetic parameters, common effect of family and genetic and phenotypic correlations for performance traits ( daily weight gain - GPD and total weight ) and yield (area, volume, body density and fillet yield ) of three genetic groups of tilapia : GIFT, ½ GIFT ½ Chitralada, 1/2 GIFT ½ Bouaké. The data set contained information from 859 animals being 563 males and 296 females. Breeding values of 968 animals in the relationship matrix were predicted. Single-trait analyzes of all traits and bicaracter associating trait fillet yield with daily weight gain, area, volume and density were performed. The analyzes were performed from the Animal Model using Bayesian inference by applying the MTGSAM program (Multiple Trait Gibbs Sampler using in Animal Model). The convergence of chains was tested by the method of Heidelberger and Welch. The heritability estimates for all traits, both in single-trait analyzes as bicaracter varied from 0.09 to 0.20 for fillet yield, indicating moderate magnitude. For the association between fillet yield with GPD, the value of the genetic correlation was -0.24 , with credibility intervals containing the zero value, showing the probability of no genetic association between these traits. Similar results were found for the genetic correlations between fillet yield and other traits, indicating the necessity of conducting a specific program of breeding for fillet yield.

Saved in:
Bibliographic Details
Main Author: Kunita, N. M.
Format: Thesis/Dissertation biblioteca
Language:Portuguese
Published: Universidade Estadual de Maringá. Centro de Ciências Agrárias. Programa de Pós-Graduação em Zootecnia 2014
Subjects:Brazil, Nile tilapia, Brasil, Gift, Fillet yield, Tilápias do Nilo, Chitralada, Rendimento de filé, Bouaké,
Online Access:http://hdl.handle.net/1834/9827
Tags: Add Tag
No Tags, Be the first to tag this record!