In lauzikaite/massFlowR: LC-MS Data Pre-Processing

knitr::opts_chunk$set(echo = TRUE,
                      collapse = TRUE,
                      eval = FALSE, 
                      message =  TRUE)
massFlowR_url <- "https://htmlpreview.github.io/?https://github.com/lauzikaite/massFlowR/blob/master/vignettes/massFlowR.html"
processing_url <- "https://htmlpreview.github.io/?https://github.com/lauzikaite/massFlowR/blob/master/vignettes/processing.html"
annotation_url <- "https://htmlpreview.github.io/?https://github.com/lauzikaite/massFlowR/blob/master/vignettes/annotation.html"

massflowR

Package for pre-processing of high-resolution, untargeted, centroid LC-MS data.

massFlowR detects and aligns structurally-related spectral peaks across LC-MS experiment samples.

Individual samples processing

Each LC-MS file in the experiment is processed independently.

Chromatographic peak detection enabled by the centWave algorithm from xcms package.
Grouping structurally-related chromatographic peaks into pseudo chemical spectra (PCS).

Individual samples can be processed in real-time during LC-MS spectra acquisition, or in parallel, using backend provided by doParallel package.

Peak alignment

To align chromatographic peaks across all samples in LC-MS experiment, an algorithm, which compares the overall similarity of pseudo chemical spectra is implemented (see Peak alignment).

Post-alignment processing

Once pseudo chemical spectra are aligned across samples, the obtained features are validated. Intensity values for each feature in a PCS are correlated across all samples to identify sets of features with similar correlation pattern.

Final step in the pipeline is to re-integrate intensity values for chromatographic peaks that were not detected by the centWave using raw LC-MS files. m/z and rt integration regions are estimated for each sample separetely through cubic smoothing spline interpolation.

In the generated feature table, each feature is assigned to a pseudo chemical spectra, which can be used in manual metabolite annotations.

Automatic annotation

If in-house chemical reference database is available, obtained PCS are annotated. For more details how to build a database file, see annotation using database).

Vignettes

More information is available in vignettes:

Installation

# To install dependencies
install_dependencies <- function () {
  dependencies <- c("xcms",  "MSnbase", "faahKO", "igraph", "doParallel",  "foreach", "ggplot2", "viridis", "gridExtra")
  installed <- installed.packages()
  to_install <- subset(dependencies, !(dependencies %in% installed[, "Package"]))
  if (length(to_install) != 0) {
    if (!requireNamespace("BiocManager", quietly = TRUE)) {
      install.packages("BiocManager")
    }
    message("Installing packages: ", to_install, " ...")
    BiocManager::install(to_install)
  } else {
    message("All dependencies are already installed.")
  }
}
install_dependencies()

# To install development version
devtools::install_github("lauzikaite/massflowR", dependencies = FALSE)