R scripts for converting & processing HPLC-MS lipid data. These scripts were developed in the Van Mooy Lab at Woods Hole Oceanographic Institution. Many of them were written specifically to prepare HPLC-ESI-MS data from an Exactive Orbitrap mass spectrometer for follow-on analysis with the LOBSTAHS package.
-
prepOrbidata.R: LOBSTAHS presumes data have been pre-processed with xcms and CAMERA; prepOrbidata.R can help with this. If the user chooses, prepOrbidata.R applies IPO to optimize xcms peak picking and retention time correction parameters. Includes a function to write the results of each IPO optimization run to a .csv file so the results aren't lost.
-
optim_centWaveParams_standalone.R: Uses IPO to optimize parameters for the xcms peak picking function centWave. Works fine on its own, but has been folded into prepOrbidata.R. Includes a function to write the results of each IPO optimization run to a .csv file so the results aren't lost.
-
Exactive_full_scan_process_ms1 .r: Scripted calling of the msconvert command line tool to convert Thermo Exactive Orbitrap .raw files obtained in polarity switching mode to .mzXML files that be read by xcms, MAVEN, mzR, or your favorite program/package. Exactive_full_scan_process_ms1 .r creates centroids and extracts positive and negative ion mode scans into separate .mzXML files. Gives user the option to convert all files in a particular directory, or only those files in a .txt document:
- mzxml_convert_list.txt: For (optional) use with Exactive_full_scan_process_ms1 .r
The raw data files that used to reside here have been incorporated into their own R package PtH2O2lipids and thus have their own repository: https://github.com/vanmooylipidomics/PtH2O2lipids
While the scripts here use xcms and CAMERA to accomplish much of the necessary processing (peak picking, retention time correction, grouping, etc.), we've also used the fast and very good program MAVEN for many projects. In fact, the older version of our analysis pipeline relied on MAVEN for peak picking, chromatographic alignment, grouping, and assignment of compound identities from our databases; the scripts for the pipeline are still in the repository https://github.com/vanmooylipidomics/old_pipeline.