Meta-análise do Projeto Toxicogenômico Japonês: diferenças entre modelos in vivo e in vitro
Carregando...
Data
2017-07-14
Autores
Biagi Júnior, Carlos Alberto Oliveira de [UNESP]
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Estadual Paulista (Unesp)
Resumo
A toxicogenômica é um campo emergente que possibilita o estudo dos efeitos de uma determinada droga a nível molecular em sistemas modelos. Uma das principais questões é se podemos substituir os estudos in vivo pelos estudos in vitro. As ciências ômicas possibilitam a resposta para esse tipo de questionamento pois fornecem técnicas como, por exemplo, o microarray, que permite o conhecimento dos transcritos (RNAs) de um dado organismo. O Projeto Toxicogenômico Japonês fornece dados para Homo sapiens, com somente experimentos in vitro, e para Rattus norvegicus, com experimentos in vitro e in vivo, tratados com 131 drogas (aprovadas pelo FDA) em diferentes concentrações de dose e tempos de amostragem, totalizando, aproximadamente, 20000 chips de microarray. A partir desses dados foi possível responder a questão inicial e observar as diferenças existentes entre cada modelo. Por meio da linguagem de programação R normalizamos os dados, obtemos os genes diferencialmente expressos e o respectivo enriquecimento funcional, dessa forma observamos as diferenças entre cada modelo. Em seguida realizamos uma análise comparativa dos modelos in vivo e in vitro adaptando a metodologia do mapa modular proposta por Segal e colaboradores. Essa metodologia tem como objetivo principal obter módulos, que são sets de genes (dados do Gene Ontology, KEGG e Reactome) que agem em conjunto para realizar uma função específica. Além de extrair módulos caracterizaremos os valores de expressão em relação as condições clínicas fornecidas pelo Projeto Toxicogenômico Japonês. Com base nessas informações do mapa modular foi possível identificar quais condições estão enriquecidas para um determinado conjunto de sets de genes, ou seja, quais processos biológico ou rotas metabólicas estão alteradas em condições específicas. Neste trabalho foi possível identificar diferenças entre os modelos in vitro e in vivo para Homo sapiens e Rattus norvegicus por meio da metodologia do mapa modular, avaliamos a quantidade de genes diferencialmente expressos e o enriquecimento funcional para diferentes concentrações de dose e diferentes tempos de amostragem. Concluímos que não é possível substituir os estudos in vivo pelo in vitro a partir dos dados analisados.
Toxicogenomics is an emerging field that allows the study of the effects of a given drug at the molecular level in model systems. One of the key issues is whether we can replace in vivo studies with in vitro studies. The omics sciences enable researchers to address this problem because it provides techniques, for example the microarray, that allow the knowledge of the transcripts (RNAs) of a given organism. The Japanese Toxicogenomic Project provides data for Homo sapiens, with only in vitro experiments, and for Rattus norvegicus, with in vitro and in vivo experiments, treated with 131 drugs (FDA approved) at different dose concentrations and sampling times, totaling approximately 20,000 microarray chips. From these data it was possible to answer the initial question and to observe the differences between each model. Using the programming language R we normalized the data, obtained the differentially expressed genes and their respective functional enrichment, so that we could observe the differences between each model. We then performed a comparative analysis of the in vivo and in vitro models by adapting the modular map methodology proposed by Segal et al. The main objective of this methodology is to obtain modules, which are gene sets (Gene Ontology, KEGG and Reactome data) that act together to perform a specific function. In addition to extracting modules we will characterize the expression values in relation to the clinical conditions provided by the Japanese Toxicogenomic Project. Based on the information provided by the modular map it was possible to identify which conditions are enriched for a given set of genes, or in other words, what biological processes or metabolic pathways are altered under specific conditions. In this work it was possible to identify differences between in vitro and in vivo models for Homo sapiens and Rattus norvegicus using the modular map methodology, we evaluated the number of differentially expressed genes and the functional enrichment for different dose concentrations and different sampling times. We conclude that it is not possible to replace the in vivo studies by in vitro from the analyzed data.
Toxicogenomics is an emerging field that allows the study of the effects of a given drug at the molecular level in model systems. One of the key issues is whether we can replace in vivo studies with in vitro studies. The omics sciences enable researchers to address this problem because it provides techniques, for example the microarray, that allow the knowledge of the transcripts (RNAs) of a given organism. The Japanese Toxicogenomic Project provides data for Homo sapiens, with only in vitro experiments, and for Rattus norvegicus, with in vitro and in vivo experiments, treated with 131 drugs (FDA approved) at different dose concentrations and sampling times, totaling approximately 20,000 microarray chips. From these data it was possible to answer the initial question and to observe the differences between each model. Using the programming language R we normalized the data, obtained the differentially expressed genes and their respective functional enrichment, so that we could observe the differences between each model. We then performed a comparative analysis of the in vivo and in vitro models by adapting the modular map methodology proposed by Segal et al. The main objective of this methodology is to obtain modules, which are gene sets (Gene Ontology, KEGG and Reactome data) that act together to perform a specific function. In addition to extracting modules we will characterize the expression values in relation to the clinical conditions provided by the Japanese Toxicogenomic Project. Based on the information provided by the modular map it was possible to identify which conditions are enriched for a given set of genes, or in other words, what biological processes or metabolic pathways are altered under specific conditions. In this work it was possible to identify differences between in vitro and in vivo models for Homo sapiens and Rattus norvegicus using the modular map methodology, we evaluated the number of differentially expressed genes and the functional enrichment for different dose concentrations and different sampling times. We conclude that it is not possible to replace the in vivo studies by in vitro from the analyzed data.
Descrição
Palavras-chave
Bioinformática, Toxicogenômica, Microarranjo