Srivastava Sudhir, Merchant Michael, Rai Anil, Rai Shesh N
Centre for Agricultural Bioinformatics, ICAR-Indian Agricultural Statistics Research Institute, New Delhi, India.
Department of Bioinformatics & Biostatistics, University of Louisville, Louisville, Kentucky, United States of America.
J Proteomics Bioinform. 2019;12(4):85-88. Epub 2019 May 23.
The proteomics experiments involve several steps and there are many choices available for each step in the workflow. Therefore, standardization of proteomics workflow is an essential task for design of proteomics experiments. However, there are challenges associated with the quantitative measurements based on liquid chromatography-mass spectrometry such as heterogeneity due to technical variability and missing values.
We introduce a web application, Proteomics Workflow Standardization Tool (PWST) to standardize the proteomics workflow. The tool will be helpful in deciding the most suitable choice for each step of the experimentation. This is based on identifying steps/choices with least variability such as comparing Coefficient of Variation (CV). We demonstrate the tool on data with categorical and continuous variables. We have used the special cases of general linear model, analysis of covariance and analysis of variance with fixed effects to study the effects due to various sources of variability. We have provided various options that will aid in finding the contribution of sum of squares for each variable and the CV. The user can analyze the data variability at protein and peptide level even in the presence of missing values.
The source code for "PWST" is written in R and implemented as shiny web application that can be accessed freely from https://ulbbf.shinyapps.io/pwst/.
蛋白质组学实验涉及多个步骤,并且在工作流程的每个步骤都有多种选择。因此,蛋白质组学工作流程的标准化是蛋白质组学实验设计的一项重要任务。然而,基于液相色谱 - 质谱的定量测量存在挑战,例如由于技术变异性导致的异质性和缺失值。
我们引入了一个网络应用程序,蛋白质组学工作流程标准化工具(PWST)来标准化蛋白质组学工作流程。该工具将有助于为实验的每个步骤确定最合适的选择。这是基于识别变异性最小的步骤/选择,例如比较变异系数(CV)。我们在具有分类变量和连续变量的数据上演示了该工具。我们使用了一般线性模型、协方差分析和固定效应方差分析的特殊情况来研究各种变异性来源的影响。我们提供了各种选项,有助于找到每个变量的平方和贡献以及CV。即使存在缺失值,用户也可以在蛋白质和肽水平分析数据变异性。
“PWST”的源代码用R编写,并实现为一个闪亮的网络应用程序,可从https://ulbbf.shinyapps.io/pwst/免费访问。