Lipidomics Workflow (v1.1.0)

Workflow Overview

The liquid chromatography-mass spectrometry (LC-MS)-based lipidomics workflow (part of MetaMS) is built using PNNL’s CoreMS software framework. The workflow leverages many features of CoreMS as well as PNNL’s MetabRef LC-MS database to process LC-MS/MS data and identify lipids. The initial signal processing includes peak picking, integration of mass features, deconvolution of MS1 spectra, and calculation of peak shape metrics. The workflow associates MS1 spectra with their corresponding MS2 spectra. It uses the MS2 spectra to search an in-silico spectra database for lipids and uses the deconvoluted MS1 spectra to assign a molecular formula. Each candidate lipid assignment is given two confidence scores: one for its match to the predicted molecular formula based on the mass accuracy and fine isotopic structure and a second for the MS2 spectral matching for filtering and selecting the best match.

Note that only data collected in profile mode for MS1 and data-dependent acquisition for MS2 is supported at this time.

Workflow Availability

The workflow is available in GitHub: https://github.com/microbiomedata/metaMS/blob/master/wdl/metaMS_lcmslipidomics.wdl

The container is available at Docker Hub (microbiomedata/metaMS): https://hub.docker.com/r/microbiomedata/metams

The database is available by request. Please contact NMDC (support@microbiomedata.org) for access.

Requirements for Execution

The recommended way to run the workflow is via the provided wdl file and the miniwdl package. Using the wdl file requires the following:

Hardware Requirements

To run this application, we recommend a processor with at least 2.0 GHz speed, 8GB of RAM, 10GB of free hard disk space.

Software Requirements

Docker Container Runtime
miniwdl (v1, https://pypi.org/project/miniwdl/)

Note that the wdl file will automatically pull the necessary docker with the required workflow dependencies.

Database

A local copy of the in silico lipid spectra. See PNNL Metabref Database (https://metabref.emsl.pnnl.gov/).

The in-silico lipid spectra in PNNL’s Metaref database are generated from the LipidBlast database (v68), found at https://systemsomicslab.github.io/compms/msdial/main.html. Note that there is no retention time in the PNNL version of the database and the workflow does not use retention time scoring. Currently the workflow uses a local copy of the database, and the database is available by request.

Sample datasets

An example dataset can be downloaded from here: https://nmdcdemo.emsl.pnnl.gov/lipidomics/blanchard_11_8ws97026/Blanch_Nat_Lip_H_32_AB_O_19_NEG_25Jan18_Brandi-WCSH5801.raw
Example CoreMS Parameter file (applicable to the example dataset): https://nmdcdemo.emsl.pnnl.gov/lipidomics/parameter_files/emsl_lipidomics_corems_params.toml
Example Scan Translator file (applicable to the example dataset): https://nmdcdemo.emsl.pnnl.gov/lipidomics/parameter_files/emsl_lipidomics_scan_translator.toml

Execution Details

This workflow should be executed using the wdl file provided in the MetaMS package (wdl/metaMS_lipidomics.wdl).

Example command to run the workflow:

miniwdl run wdl/metaMS_lipidomics.wdl -i metams_input_lipidomics.json --verbose --no-cache --copy-input-files

Inputs

To use the wdl, inputs should be specified in a json file. See example input json file in wdl/metaMS_lipidomics.wdl.

The following inputs are required (declared in the input json file):

LC-MS/MS data file locations in one of the following formats:
- ThermoFisher mass spectrometry data files (.raw)
- mzML mass spectrometry data files (.mzml)
Workflow inputs (all three required):
- CoreMS Parameter file (.toml)
- Scan Translator Parameter file (.toml)
- Path to local MetabRef database (.sqlite)
Cores (optional input):
- How many cores to use for processing. Default is 1.

Outputs

Lipidomics data - Peak data table with annotated lipids (.csv) - CoreMS HDF5 format of CoreMS LCMS object for further analysis and exploration (.hdf5)
Workflow Metadata - The full set of CoreMS parameters used in the workflow, some of which are set dynamically within the workflow (.toml)

Version History

v1.1.0: Initial release of the lipidomics workflow Jan 9, 2025

Point of contact

Workflow maintainer: Katherine R. Heal <katherine.heal@pnnl.gov>