Application of neural networks and random regression models for predict the breeding value of milk production in Holstein cows

• Grant number: 12/16087-5
• Support type: Scholarships abroad - Research Internship - Doctorate
• Effective date (Start): January 02, 2013
• Effective date (End): August 01, 2013
• Field of knowledge: Agronomical Sciences - Animal Husbandry - Genetics and Improvement of Domestic Animals
• Principal researcher: Danísio Prado Munari
• Grantee: Rodrigo Pelicioni Savegnago
• Supervisor abroad: Guilherme J. M. Rosa
• Home Institution: Faculdade de Ciências Agrárias e Veterinárias (FCAV). Universidade Estadual Paulista (UNESP). Campus de Jaboticabal. Jaboticabal , SP, Brazil
• Research place: University of Wisconsin-Madison (UW-Madison), United States
• Associated to the scholarship: 10/05148-8 - Application of neural networks and random regression models for predict the breeding value of milk production in Holstein cows, BP.DR

Abstract

The aim of this study is to evaluate the potential of the use of the artificial neural networks to predict the milk production and the estimates of breeding values of this trait in Holstein cows. It will be evaluated about 30,000 monthly dairy milk controls of approximately 3,500 cows in the first lactation. It will be used a random regression model to estimate the genetic parameters of the monthly milk production. The fixed effects of the model will be contemporary groups, the linear and quadratic effects of age off calving and the mean curve off lactation and the direct additive genetic, permanent environment and residual random effects. Multilayer Perceptron neural networks models (MLP) will be evaluated. The milk production of collaterals and/or ancestors, the breeding value the animals and/or the parents and the monthly milk production will be used in the input layer of neural networks to predict the total milk production and the breeding values of this trait, used in the output layer of the MLP neural network. The correlation between the estimates of breeding values of milk production obtained by the MLP models with the estimated of breeding values of milk production obtained by the random regression models (control values) will be used to evaluate the performance of the neural networks to obtain this estimates. (AU)

Scientific publications

(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)

DE FREITAS, LUARA AFONSO; SAVEGNAGO, RODRIGO PELICIONI; GRUPIONI, NATALIA VINHAL; RAMOS, SALVADOR BOCCALETTI; STAFUZZA, NEDENIA BONVINO; PEREIRA DE FIGUEIREDO, ELSIO ANTONIO; SCHMIDT, GILBERTO SILBER; LEDUR, MONICA CORREA; MUNARI, DANISIO PRADO. Reduced-rank estimation of genetic parameters for egg production traits and cluster analyses with predicted breeding values. ACTA AGRICULTURAE SCANDINAVICA SECTION A-ANIMAL SCIENCE, v. 68, n. 2, p. 81-86, APR 3 2018. Web of Science Citations: 0.

SAVEGNAGO, RODRIGO PELICIONI; DO NASCIMENTO, GUILHERME BATISTA; DE MAGALHAES ROSA, GUILHERME JORDAO; RESENDE DE CARNEIRO, RAUL LARA; SESANA, ROBERTA CRISTINA; EL FARO, LENIRA; MUNARI, DANISIO PRADO. Cluster analyses to explore the genetic curve pattern for milk yield of Holstein. LIVESTOCK SCIENCE, v. 183, p. 28-32, JAN 2016. Web of Science Citations: 4.

SAVEGNAGO, R. P.; ROSA, G. J. M.; VALENTE, B. D.; HERRERA, L. G. G.; CARNEIRO, R. L. R.; SESANA, R. C.; EL FARO, L.; MUNARI, D. P. Estimates of genetic parameters and eigenvector indices for milk production of Holstein cows. JOURNAL OF DAIRY SCIENCE, v. 96, n. 11, p. 7284-7293, NOV 2013. Web of Science Citations: 5.