Modelos estocásticos de transmissão para análises genéticas de características epidemiológicas

• Ano de defesa: 2023
• Autor(a) principal: Lima, Milena Nascimento
• Orientador(a): Anacleto Junior, Osvaldo
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de São Carlos; Câmpus São Carlos
• Programa de Pós-Graduação: Programa Interinstitucional de Pós-Graduação em Estatística - PIPGEs
• Palavras-chave em Português: Doenças infecciosas; Susceptibilidade; Infectividade; Genética quantitativa; Inferência Bayesiana
• Palavras-chave em Inglês: Infectious diseases; Susceptibility; Infectivity; Quantitative genetics; Bayesian inference
• Área do conhecimento CNPq: CIENCIAS EXATAS E DA TERRA::PROBABILIDADE E ESTATISTICA::ESTATISTICA::INFERENCIA EM PROCESSOS ESTOCASTICOS
• Link de acesso: https://repositorio.ufscar.br/handle/20.500.14289/18683

Epidemics can significantly affect animal production and generate large economic impacts. Furthermore, current practices for treating and controlling infectious diseases in farmed animals do not always show the desired effectiveness. In these cases, the quantitative genetics offers a viable alternative through the study of genetic variations of host characteristics that affect disease transmission, especially susceptibility, and infectivity. However, despite advances, the challenge to estimate genetic effects that mainly control infectivity continues to exist. Therefore, the general objective of this thesis was to contribute to the state of the art in the development of statistical models that can capture the dynamics of transmission of infectious diseases and consequently improve the estimation of genetic effects on infectivity. We present a new version of the dynamic non-linear indirect genetic effects model (dnIGE) and an inferential method to estimate its parameters. Our methodology includes a covariance structure on the distribution of genetic and environmental effects of susceptibility and infectivity, which were previously considered independent and uses modern Bayesian inference to estimate the genetic effects and heritabilities associated with these traits. Results show that the extended dnIGE model can accurately estimate heritabilities and genetic values associated with susceptibility and infectivity, even when there is a genetic correlation between these traits. Our proposed methodology offers potential impacts in areas such as disease control in livestock through selective breeding and also in predicting and controlling the emergence of disease outbreaks in human populations.