Lemoine Sophie, Combes Florence, Servant Nicolas, Le Crom Stéphane
IFR36, Plate-forme Transcriptome, Ecole Normale Supérieure, 46 rue d'Ulm, 75230 Paris cedex05, France.
BMC Bioinformatics. 2006 Oct 23;7:467. doi: 10.1186/1471-2105-7-467.
Raw data normalization is a critical step in microarray data analysis because it directly affects data interpretation. Most of the normalization methods currently used are included in the R/BioConductor packages but it is often difficult to identify the most appropriate method. Furthermore, the use of R commands for functions and graphics can introduce mistakes that are difficult to trace. We present here a script written in R that provides a flexible means of access to and monitoring of data normalization for two-color microarrays. This script combines the power of BioConductor and R analysis functions and reduces the amount of R programming required.
Goulphar was developed in and runs using the R language and environment. It combines and extends functions found in BioConductor packages (limma and marray) to correct for dye biases and spatial artifacts. Goulphar provides a wide range of optional and customizable filters for excluding incorrect signals during the pre-processing step. It displays informative output plots, enabling the user to monitor the normalization process, and helps adapt the normalization method appropriately to the data. All these analyses and graphical outputs are presented in a single PDF report.
Goulphar provides simple, rapid access to the power of the R/BioConductor statistical analysis packages, with precise control and visualization of the results obtained. Complete documentation, examples and online forms for setting script parameters are available from http://transcriptome.ens.fr/goulphar/.
原始数据归一化是微阵列数据分析中的关键步骤，因为它直接影响数据解读。目前使用的大多数归一化方法都包含在R/BioConductor软件包中，但通常很难确定最合适的方法。此外，使用R命令进行函数和图形操作可能会引入难以追踪的错误。我们在此展示一个用R编写的脚本，它为双色微阵列的数据归一化提供了一种灵活的访问和监测方式。该脚本结合了BioConductor和R分析功能的强大之处，并减少了所需的R编程量。
Goulphar是用R语言和环境开发并运行的。它结合并扩展了BioConductor软件包（limma和marray）中的功能，以校正染料偏差和空间伪影。Goulphar在预处理步骤中提供了广泛的可选和可定制过滤器，用于排除不正确的信号。它显示信息丰富的输出图，使用户能够监测归一化过程，并有助于使归一化方法适当地适应数据。所有这些分析和图形输出都呈现在一份单一的PDF报告中。
Goulphar提供了对R/BioConductor统计分析软件包强大功能的简单、快速访问，对所得结果进行精确控制和可视化。完整的文档、示例以及用于设置脚本参数的在线表单可从http://transcriptome.ens.fr/goulphar/获取。
