Calico Lipidomics and Metabolomics Analysis
Claman provides methods for analyzing mzroll metabolomics/lipidomics datasets generated using MAVEN.
Each mzroll (.mzrollDB) file is an sqlite database that represents a single experimental metabolomics or lipidomics experiment generated using a shared analytical method (same mode, fragmentation, chromatography, etc). The major experimental data used for interpreting an LC-MS experiment are abundances of peaks across samples for each distinct peakgroup (ions with a characteristic mass and retention time). Claman extracts this data from mzrolls and stores it as an mzroll_list; a list of peakgroups, samples, and peaks. This data is further augmented with feature-attributes such as pathway information and with sample-attributes which reflect an experiments design. The addition of sample attributes, allows users to integrate mzrolls which include a common set of samples that were profiled using different methods (such a positive-, and negative-mode measurements).
mzroll_lists can then be filtered, floored to the limit of detection, and normalized using a suite of built-in methods that can correct for differences in sample loading, compare treatments to a reference condition, and align batches. Next, analytes of interest can be identified using peakgroup-level statistical analysis. Significant findings can then be explored using peakgroup-level visualizations and by identifying pathways that are enriched for significant metabolite changes.
The package can be installed from GitHub using:
# install.packages("remotes")
remotes::install_github('calico/claman', dependencies = TRUE)
# for vignettes
remotes::install_github(
'calico/claman',
dependencies = TRUE,
build_vignettes = TRUE
)
After installing this package, the package functions can be accessed by just loading the package in R:
library(claman)
Once claman is installed, run package vignette(s):
vignette(package = "claman", topic = "NPLUG")
And, check out claman's pkgdown site for organized documentation.