O microbioma humano como indicador de saúde: correlações e ferramentas analíticas

• Ano de defesa: 2021
• Autor(a) principal: Dobbler , Priscila Caroline Thiago
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal do Pampa; UNIPAMPA; Doutorado em Ciências Biológicas; Brasil; Campus São Gabriel
• Palavras-chave em Português: 16S; Diversidade microbiana; Sequenciamento de nova geração; Gravidez; Core microbioma; Bioinformática; Microbial diversity; Next Generation sequencing; Pregnancy; Core Microbiome; CNPQ::CIENCIAS BIOLOGICAS
• Link de acesso: https://repositorio.unipampa.edu.br/jspui/handle/riu/7207

For a while, microorganisms were considered harmful to the health and normal functioning of the human body, and the pathologies associated with these organisms were the chief objective of studies. With the evolution of the microbiology field, and more recently, the development and expansion of sequencing technologies, microorganisms came to be seen under a new perspective. The advent of sequencing for cataloging microbial communities has made possible to identify hundreds to thousands of microorganisms in a single sample. This highlighted the interdependency of humans with the different types of microbial communities. Therefore, microbiome research seeks out to identify correlations, not only with microorganisms in isolation, but also at community level, where there are altered correlations, or co-occurrences, that lead to community metabolic unbalancing that reflects on the host. Furthermore, pregnant women harbor characteristic vaginal microbial communities that are distinct from that of non-pregnant women. Also, during a healthy pregnancy, these communities also undergo important transformations. During a vaginal delivery, the baby goes through the birth canal and gets in contact with the local microbiota. Considering this, we sought to understand how the maternal vaginal microbiota, at the end of pregnancy, is associated with the newborn’s microbiota. When carrying out this study, we realized that the tools available were not adequate to deal with the intragroup variability, or data sparsity. For this reason, we created and validated a tool called PIME (Prevalence Interval for Microbiome Evaluation) that is capable of reduce intragroup variability by using varying levels of prevalence for filtering. PIME is also capable of classifying the most important taxonomic units for differentiating between groups, by using Random Forests algorithm.