Introduction

When using Biocrates metabolomic panels, data is managed using the MetIDQ software. For data analysis in R is used excel files exported from MetIDQ without normalization.

This package provides:

• an import function that parses said excel file into a Biobase::ExpressionSet
• functions for preprocessing and visualization of omics data, that work with ExpressionSets such as:
  • plot_layout to plot the arrangement of sample on the plate
  • plot_pca to plot a PCA with one grouping factor
  • plot_qcs to plot the expression levels of selected metabolites over processing time (to investigate any drift during sample measurement) while connecting the measurements of QC-samples
  • plot_features to produce boxplots of each feature
  • aggregate_eset to easily merge two ExpressionSets
  • qc_rlsc to perform QC-RLSC normalization on a selected QC-sample
  • and many more...

Details

You can learn more about the Biobase::ExpressionSet here. This data structure, initially developed for transcriptomic assays, ensures that the measured feature expression (e.g. in this case metabolite concentrations) stays with the sample data and any additional information like the LOD for each metabolite and batch. Filtering an ExpressionSet by sample data is easy and prevents getting mixed up between several tables, potentially not ordered accurately.

Licence

The package was written for my personal use and is still work in progress (bugs are to be expected)! It is provided under a CC-BY-SA 4.0 licence.

Installation

This R package can be installed via github:

BiocManager::install("Biobase")
devtools::install_github("jkdel/IDQprocessor")