Development of statistical skills using machine learning to evaluate sheep feed and water efficiency data

• Grant number: 23/04641-2
• Support Opportunities: Scholarships abroad - Research Internship - Post-doctor
• Start date: July 01, 2023
• End date: July 31, 2023
• Field of knowledge: Agronomical Sciences - Animal Husbandry - Animal Production
• Principal Investigator: Ricardo Lopes Dias da Costa
• Grantee: Josiel Borges Ferreira
• Supervisor: Luis Alberto Bermejo Asensio
• Host Institution: Instituto de Zootecnia. Agência Paulista de Tecnologia dos Agronegócios (APTA). Secretaria de Agricultura e Abastecimento (São Paulo - Estado). Nova Odessa , SP, Brazil
• Institution abroad: Universidad de La Laguna (ULL), Spain
• Associated to the scholarship: 22/10795-0 - Growth efficiency, passage rate, methane emission, carcass and meat characteristics and their correlations with thermography in ewes and lambs selected for residual feed intake (CAR), BP.PD

Abstract

The phenotypic evaluation of feed efficiency traits for ruminant species through feed efficiency tests is currently common. The use of this information helps genetic improvement programs, bringing great benefits to production systems, as well as to planet. The large amount of data produced from feed efficiency tests is usually analyzed in a "univariate" way. However, thinking in a global context of database evaluation, large databases could be analyzed in a multivariate way (machine learning), elucidating the relationships that phenotypic estimates of feed efficiency establish with each other to determine and select efficient animals in conversion of protein feeds, or even efficient in water consumption, in addition to defining potentially useful variables as a phenotypic marker of feed efficiency in sheep. The aim of this project is to provide the development of statistical skills based on machine learning techniques for data analysis from water and feed efficiency tests for sheep and subsequent genetic improvement. This proposal includes the post-doctoral project of the candidate for Bolsa Estágio de Pesquisa no Exterior (BEPE), entitled "Eficiência de crescimento, taxa de passagem, emissão de metano, características de carcaça e carne e suas correlações com a termografia em ovinos e cordeiros selecionados para consumo alimentar residual (CAR) - FAPESP - 2022/10795-0". Data collection and laboratory analyzes (feed and water efficiency tests) were carried out between November 2022 and February 2023, therefore, there is a need to start statistical analyses. With the use of machine learning techniques, it is expected to identify the main relationships between the variables of feed and water efficiency for sheep, defining by degrees of importance the main variables linked to feed efficiency that could potentially be used as phenotypic markers to select animals efficient and support genetic improvement programs. This proposal will be directly linked to the proponent's research project and indirectly to the statistical analyzes of master's work by students of the Programa de Pós-Graduação em Produção Animal Sustentável at the Instituto de Zootecnia, which are in progress or have already been completed, supporting the application of new statistical analysis tools and enabling and creating new associated projects. (AU)