• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 1
  • Tagged with
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Simula??o em n?vel de gene e de indiv?duo aplicada ao melhoramento animal / Simulation of individual and gene level applied to animal breeding

Farah, Michel Marques 15 July 2010 (has links)
Submitted by Rodrigo Martins Cruz (rodrigo.cruz@ufvjm.edu.br) on 2015-12-17T16:32:54Z No. of bitstreams: 2 michel_marques_farah.pdf: 437133 bytes, checksum: efc5c1b8937d6edbcc7aaf0f1481a293 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Approved for entry into archive by Rodrigo Martins Cruz (rodrigo.cruz@ufvjm.edu.br) on 2015-12-17T16:33:17Z (GMT) No. of bitstreams: 2 michel_marques_farah.pdf: 437133 bytes, checksum: efc5c1b8937d6edbcc7aaf0f1481a293 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) / Made available in DSpace on 2015-12-17T16:33:19Z (GMT). No. of bitstreams: 2 michel_marques_farah.pdf: 437133 bytes, checksum: efc5c1b8937d6edbcc7aaf0f1481a293 (MD5) license_rdf: 0 bytes, checksum: d41d8cd98f00b204e9800998ecf8427e (MD5) Previous issue date: 2010 / Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior (CAPES) / A simula??o de dados apresenta diversas vantagens, como proporcionar a obten??o de respostas ? sele??o e diminuir o tempo necess?rio para a avalia??o das metodologias estudadas no melhoramento gen?tico animal. Por?m, os trabalhos que utilizam simula??o empregam v?rios termos como simula??o estoc?stica, simula??o determin?stica, simula??o de Monte Carlo, simula??o em n?vel de gene e simula??o em n?vel de indiv?duo e, muitas vezes, estes termos s?o utilizados de maneiras diferentes ou em outras condi??es, causando uma diverg?ncia nos termos utilizados. Assim, os objetivos deste trabalho foram agrupar, definir e diferenciar os termos t?cnicos utilizados nos trabalhos de simula??o em melhoramento gen?tico animal e comparar e definir as propriedades dos procedimentos de simula??o em n?vel de indiv?duo e em n?vel de gene. Foram desenvolvidos tr?s cen?rios de simula??o, em n?vel de indiv?duo, em n?vel de gene com e sem marcador utilizando o software LZ5. Foram simuladas tr?s popula??es de su?nos para cada cen?rio e com diferentes herdabilidades (0,12, 0,27 e 0,47). A popula??o-base foi constitu?da de 1500 animais, sendo 750 machos e 750 f?meas e para as duas simula??es em n?vel de gene foi considerado um genoma de 2800 cM e 18 cromossomos de tamanhos aleat?rios, as caracter?sticas foram governadas por 500 locos polig?nicos dial?licos, com freq??ncias al?licas iguais e taxa de recombina??o de 0,01. Para a simula??o em n?vel de gene com marcadores, ainda foram distribu?dos marcadores distanciados igualmente a 50 cM e distribu?dos aleatoriamente 5 QTLs por todo o genoma. Os valores amostrados apresentaram bem semelhantes para os tr?s tipos de simula??o, apresentando um aumento das vari?ncias aditiva e fenot?pica e da herdabilidade nas primeiras gera??es e depois decrescendo ao longo das gera??es. J? para a m?dia fenot?pica, houve um ganho gen?tico por gera??o, indicando que todos os m?todos utilizados s?o eficientes para a obten??o de dados simulados. Assim, a vantagem da simula??o em n?vel de gene ? que ? poss?vel simular marcadores moleculares e QTLs, enquanto a simula??o em n?vel de indiv?duo ? muito eficiente para obten??o de dados como o valor gen?tico do indiv?duo e da m?dia fenot?pica da popula??o em um per?odo de tempo muito menor, pois demanda menos recursos computacionais e de algoritmos estruturados para desenvolver quando comparado com a simula??o em n?vel de gene. Portanto, define-se simula??o em n?vel de indiv?duo como uma metodologia de simula??o que consiste em gerar valores gen?ticos (G) a partir de uma distribui??o normal com m?dia e vari?ncia previamente definidas; enquanto para a simula??o em n?vel de gene a metodologia consiste em gerar os valores dos efeitos de cada loco polig?nico e seus QTLs, a partir de uma distribui??o normal com m?dia e vari?ncia previamente definidas para cada componente, e pela soma destes, obt?m-se o G de cada indiv?duo da popula??o. Para a gera??o do efeito residual (E) as duas metodologias de simula??o s?o feitas da mesma forma, gerando-se um efeito aleat?rio amostrado, tamb?m, de uma distribui??o normal e assim obt?m-se os valores fenot?picos (P) de cada indiv?duo pela soma destes dois componentes (G+E). / Disserta??o (Mestrado) ? Programa de P?s-Gradua??o em Zootecnia, Universidade Federal dos Vales do Jequitinhonha e Mucuri, 2010. / ABSTRACT The simulation data has several advantages, such as providing the obtaining responses to selection and reduce the time required for evaluation methodologies studied in animal breeding. However, simulation studies employ various terms such as simulation stochastic, deterministic simulation, Monte Carlo simulation, simulation level of gene and simulation at the individual level and often these terms are used in different ways or in other conditions, causing a divergence in the terms used. Thus, the objectives were cluster, define and differentiate the technical terms used in the work of simulation in animal breeding and compare and define the properties procedures for simulation-level and individual-level gene. There had been developed three scenarios for simulation at the individual-level and level gene, with and without marker, using the software LZ5. There had been simulated three pig populations for each scenario, with different heritabilities (0.12, 0.27 and 0.47). The base population consisted of 1500 animals, 750 males and 750 females and for both simulations at the level of the gene was considered a genome of 2800 cM, and 18 chromosomes in random sizes, the characteristics were governed by 500 loci diallelic polygenic, with equal allele frequencies and recombination rate of 0.01. For the simulation Level with gene markers, were also distributed bookmarks equally spaced at 50 cM and five QTL distributed randomly across the genome. The sampled values were very similar for the three types of simulation, an increase of additive variance and phenotype and heritability in the first generations and then decreasing to over the generations. As for the average phenotype was a genetic per generation, indicating that all methods used are efficient for obtain simulated data. Thus, the advantage of gene-level simulation is that it can simulate molecular markers and QTLs, while the simulation at individual level is very efficient for obtaining data as the individual's genetic value and phenotypic average of the population over a period of much less time, since it requires less computational resources and algorithms structured to develop, when compared with the simulation-level gene. Therefore, it is defined as the individual level simulation a methodology simulation that generates breeding values (G) from a normal distribution with mean and variance as previously defined; and the gene level simulation is defined as a methodology that generates the values of effects of each locus and their polygenic QTLs from a normal distribution with mean and variance previously defined for each component, and the sum of these gives the G of each individual in the population. For the generation of residual effect (E) the two simulation methodologies are made in the same way, generating a random effects sampled also a normal distribution and so it was obtained the phenotypic values (P) of each individual by summing these two components (G+E).

Page generated in 0.1023 seconds