Advanced search
Start date
Betweenand

Development of a free-software to simulation in animal breeding

Grant number: 05/01554-3
Support type:Research Grants - Young Investigators Grants
Duration: December 01, 2005 - November 30, 2010
Field of knowledge:Agronomical Sciences - Animal Husbandry
Principal Investigator:Ricardo da Fonseca
Grantee:Ricardo da Fonseca
Home Institution: Universidade Estadual Paulista (UNESP). Campus Experimental de Dracena. Dracena , SP, Brazil
Assoc. researchers:Adam Taiti Harth Utsunomiya ; Aldrin Vieira Pires ; Giselle Mariano Lessa de Assis ; José Marques Carneiro Júnior ; Michele Porto Pires ; Onofre Jose Piva ; Pedro Paulo Ferreira Brito ; Rafael Keith Ono ; Rodrigo Ibelli Vaz
Associated scholarship(s):09/16993-3 - Free software development for data simulation applied to animal breeding, BP.IC

Abstract

In the animal breeding field, due to the high costs of set and maintenance of the experiments, due to large generation times and theoretical complexity, the simulation software have become very important. However, the developed programs present some limitations: 1) they are specific; 2) they present costs of purchase and/or use; 3) they are build for just one operating system; 4) they are incapable to manipulate large data sets; 5) correction and updates are slow and 6) they have not the source-code available to modification and betterments. Despite of the importance of the development of simulators to become viable experiments and other activities in animal breeding, few are the initiatives, in Brazil, of creating and maintain a research line aimed to yield software for the area. In this manner, experiments that could be simulated must be executed with high costs of setting and maintaining or they are considered not viable and are not executed. In order to the scientific community can benefit itself of the all advantages offered by simulators, those restrictions must be surpassed. A promising alternative is the development of simulators under the free-software model. Thus, the objectives of this work are: 1) develop a simulator under the free-software model to aid research in animal breeding; 2) start a research line, in Brazil, to development of free-software in animal breeding; 3) create a community of researchers who believe and share values of group work and mutual helping to development of free-software linked to animal production. It will be developed, at the same time, the genome generator, the population generator and the simple selection methods for one and two traits: mass selection, family selection and within family selection, progeny test, sibs test, tandem method and independent culling levels method. In this step, it will be implemented the functions to generate cloned and transgenic animals. Following that, the index selection will be developed and after, the prediction methods as BLUP, BLUP with identified genes and BLUP with molecular markers will be coded. Finally, the estimation of variance components and the Bayesian methods for estimation of variance components and prediction of genetic merits will be coded. The results obtained will be analyzed by the evaluation of predicted and actual genetic gains, indirect responses, inbreeding coefficients and graphics of these measures. The software will be freely distributed to universities and public institutes and will allow that several situations, with one or more modified parameters, be exposed to the students to learning the theory and practical applications. In the research field, scenarios can be simulated to verify the performance of a methodology or to compare of two or more methodologies. Assumptions of methods can be tested and preliminary results to research proposals can be simulated. Besides that, with few modifications in the program, the researcher will be able to run situations associated to specific problems, enrich scientific papers discussion, clarify uncertain points and direct the work to more promising ways. (AU)