Machine learning for predictive analysis in Santa Inês sheep: an example of application to predict resistant, resilient and susceptible animals

• Grant number: 20/03575-8
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• Start date: August 01, 2021
• End date: July 31, 2022
• Field of knowledge: Agronomical Sciences - Animal Husbandry - Genetics and Improvement of Domestic Animals
• Principal Investigator: Claudia Cristina Paro de Paz
• Grantee: Luara Afonso de Freitas Januário
• Supervisor: Guilherme Jordão de Magalhães Rosa
• 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: University of Wisconsin-Madison (UW-Madison), United States
• Associated to the scholarship: 18/01540-2 - Prediction of genomic values using Bayesian models and neural networks for endoparasite resistance traits in Santa Inês sheep, BP.DR

Abstract

Nowadays, the biggest problem for the expansion of the sheep industry is the incidence of gastrointestinal nematodes infection, which may in some situations make production systems economically unfeasible. Therefore, this study aims to compare the prediction accuracy of Multi-layer perceptron neural networks (MPL), Random Forest (RF), and Linear Discriminant Analysis (LDA) with the standard multinomial logistic regression to predict resistant, resilient and susceptible animals to endoparasites. The database contains 3,896 records of 974 Santa Ines breed animals. The phenotypic traits to be assessed as predictors for classification as resistant, resilient and susceptible will be body weight, body measurements (croup height, withers height, body length, chest girth, and croup width), and body condition score, taking into account also the environmental effects of deworming practices, farm, age, birth season, weighing season and sex. The response variable for the prediction models will be the classification of animals into resilient (1), susceptible (2) and resistant (3). The analyses will be performed using the R software. (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)

FREITAS, LUARA; SAVEGNAGO, RODRIGO; CARVALHO ALVES, ANDERSON A.; COSTA, RICARDO; ROSA, GUILHERME J. J. M.; PAZ, CLAUDIA. Classification Performance of Multinomial Logistic Regression for Identifying Resistance, Resilience, and Susceptibility to Gastrointestinal Nematode Infections in Sheep. JOURNAL OF ANIMAL SCIENCE, v. 100, p. 1-pg., 2022-09-22. (16/14522-7, 18/01540-2, 20/03575-8)

FREITAS, LUARA; FERREIRA, RAFAEL; SAVEGNAGO, RODRIGO; DOREA, JOAO R.; ROSA, GUILHERME J. J. M.; PAZ, CLAUDIA. Computer Vision System to Predict Famacha (c) Degree in Sheep from Ocular Conjunctiva Images. JOURNAL OF ANIMAL SCIENCE, v. 100, p. 1-pg., 2022-10-01. (16/14522-7, 18/01540-2, 20/03575-8)