Skip to contents

MSdev

LC-MS data analysis shell for project orchestration, xcms-based feature processing, spectra handling, annotation, and downstream analysis.

Chemistry and molecule-graph primitives live in companion package MSCC; MSdev consumes them for network / atom-transfer workflows.

Installation 

# install.packages("remotes")
remotes::install_github("DrRuiLi/MSdev")

Requires R ≥ 4.3 and Bioconductor packages such as xcms, Spectra, and ProtGenerics. Install Bioconductor dependencies first if needed:

if (!requireNamespace("BiocManager", quietly = TRUE)) {
  install.packages("BiocManager")
}
BiocManager::install(c("xcms", "Spectra", "ProtGenerics", "MsExperiment"))

What it does

Area Capabilities
Project shell MSdev S4 object: sample info, xcms data, spectra, annotation, advanced analysis
Feature processing Peak picking / grouping / RT alignment via xcms; polarity-aware MS1 containers
Spectra Extract MS1/MS2, assign MS2 to features (sp_id), on-disk storage
Feature grouping Stock xcms compounding (EicSimilarityParam) and custom EIC similarity (MSdev_group_feature_EIC)
Annotation MS2 scoring and formula / library glue
Other workflows DDA / pseudo-MS2, MRM, statistics, pathway helpers, metabolic flux network (via MSCC)

Typical untargeted workflow 

library(MSdev)

object <- MSdev(rawDataDir = "path/to/raw")

object <- MSdev_checkSampleInfo(object)
object <- MSdev_msConvert(object)
object <- MSdev_xcmsProcessing(object)
object <- MSdev_extract_Spectra(object)   # also assigns MS2 to features
object <- MSdev_group_feature_EIC(object) # optional custom EIC grouping
object <- MSdev_annotation(object)

Prefer accessors over slot digging:

  • get_MSdev_Spectra(object, msLevel = …, polarity = …)
  • get_MSdev_Chromatogram(object, polarity = …)

Documentation

Package vignettes and articles (pkgdown):

  • Get started: vignettes/MSdev.Rmd
  • Untargeted workflow notes: vignettes/MSdev_untargeted_workflow.Rmd
  • Articles under vignettes/articles/:
    • Spectra backends (MsBackend*)
    • Fast chromatogram extraction (xcms vs MSdev triad)
    • Feature grouping with EicSimilarityParam
    • Rdisop mass decomposition (decomposeMass)

Build the site locally with pkgdown when configured:

pkgdown::build_site()
Package Role
MSCC Chemistry, molecule graphs, adduct / formula helpers
xcms Peak detection, correspondence, feature compounding
Spectra Spectrum containers and backends

Author

Rui Li (ORCID) —

License

MIT © Rui Li