Carvalho Benilton S
Brazilian Institute of Neuroscience and Neurotechnology (BRAINN) and Department of Statistics, University of Campinas, Rua Sérgio Buarque de Holanda, 651 - Cidade Universitária, Campinas, São Paulo, Brazil.
Methods Mol Biol. 2016;1418:145-59. doi: 10.1007/978-1-4939-3578-9_7.
Preprocessing microarray data consists of a number of statistical procedures that convert the observed intensities into quantities that represent biological events of interest, like gene expression and allele-specific abundances. Here, we present a summary of the theory behind microarray data preprocessing for expression, whole transcriptome and SNP designs and focus on the computational protocol used to obtain processed data that will be used on downstream analyses. We describe the main features of the oligo Bioconductor package, an application designed to support oligonucleotide microarrays using the R statistical environment and the infrastructure provided by Bioconductor, allowing the researcher to handle probe-level data and interface with advanced statistical tools under a simplified framework. We demonstrate the use of the package by preprocessing data originated from three different designs.
微阵列数据预处理包括许多统计程序，这些程序将观察到的强度转换为代表感兴趣的生物学事件的量，如基因表达和等位基因特异性丰度。在这里，我们总结了用于表达、全转录组和SNP设计的微阵列数据预处理背后的理论，并重点介绍了用于获得将用于下游分析的处理后数据的计算协议。我们描述了oligo Bioconductor软件包的主要特征，该软件包是一个应用程序，旨在使用R统计环境和Bioconductor提供的基础设施来支持寡核苷酸微阵列，使研究人员能够在简化的框架下处理探针级数据并与先进的统计工具进行交互。我们通过预处理来自三种不同设计的数据来演示该软件包的使用。
