Influência do número de repetições na identificação de genes diferencialmente expressos em experimentos de RNA-Seq

• Autor(a) principal: Gonçalves, Jaciane Coelho
• Data de Publicação: 2013
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: LOCUS Repositório Institucional da UFV
• Texto Completo: http://locus.ufv.br/handle/123456789/4066
• topic: RNA-Seq; Genes diferencialmente expressos; Número de repetições; RNA-Seq; Differentially expressed genes, Number of repetitions; CNPQ::CIENCIAS AGRARIAS

One of the current objectives of molecular biology is to measure and assess the gene expression profiles in different types of biological tissues, to understand the mechanisms of molecular transformation under certain conditions. Next-generation sequencing (NGS) technologies promote DNA sequencing in platforms capable of generating information about millions of base pairs in a single step. However these technologies still have high cost, making it difficult to obtain large number of repetitions of sample data. Therefore, it becomes necessary the discovery and the improvement of efficient statistical methodologies for optimizing analysis of data generated in genome sequencing platforms. The overall objective of this work was to evaluate the effect of the number of repetitions in the identification of differentially expressed genes, in RNA-Seq experiments, contributing to the clarification of the statistic that researchers will assist in data analysis in RNA-Seq experiments. Specifically, we evaluate empirically the effect of the number of repetitions in the statistical analysis of gene expression in RNA-Seq experiments. To carry out the analyses we used a dataset defined in Li et al. (2008), which compared treated and non-treated cancer cells. That work had four biological replications for the control group (non-treated) and three replications for biological treatment group (cells that have received the treatment). The data was analyzed using the package DESeq from the statistical environment R. A total of 2566 genes were considered differentially expressed (DE) when we evaluate the original and complete dataset. When we analyzed three replications of the control and treatment, we found, on average, 2153 genes DE. From the moment in which only two reps for both treatments were used, were identified, on average, 1241 genes DE. The major change in the number of genes DE was observed when replications were not used. In this case we identified around 44 differentially expressed genes. According to the results generated in the analysis, it was possible to verify that the number of repetitions is an essential factor in order to obtain a significant number of differentially expressed genes.