No logo

OzFAD

Ozone-enabled fatty acid discovery

Description

Workflow facilitating discovery of unsaturated fatty acids through coupling liquid chromatography and mass spectrometry with gas-phase ozonolysis of double bonds. The workflow encompasses semi-automated data analysis and enables de novo identification in human, animal, cell or plant derived media. The targeted analysis including ozonolysis enables structural assignment over a dynamic range of five orders of magnitude, even in instances of incomplete chromatographic separation. Detection, without prior knowledge, allows discovery of non-canonical double bond positions. The data analysis is based on open source code, including the Skyline Mass Spectrometry Environment and custom python code - refer to the github repository noted in the associated publication.

Technical Information

Website:
https://www.nature.com/articles/s41467-023-39617-9

Covered lipid categories:
FA;

Publications:
https://www.nature.com/articles/s41467-023-39617-9

Documentation:
https://www.nature.com/articles/s41467-023-39617-9

Source code:
https://github.com/jphmenzel/jpmlipidomics

Programming languages:
Python, Batchfile

Platforms:
Skyline Mass Spectrometry Environment

Input formats:
Raw data from OzID-modified Synapt G2-Si

Output formats:
Csv, xlsx, sky

Interface:
GUI, CLI

Licence:
GLP-3.0 license

Sections

Section 3 — Targeted Lipidomics (SRM, MRM, PRM)

Analytical strategies:
Fragmentation is by ion-ion reaction of ozone with unsaturated fatty acyls, leading to diagnostic product ions. Custom and automated definition of transitions is dependent on the input raw dataset from the DIA LC-OzID-MS acquisition.

Number of transitions:
For LC-OzID-MS/MS DDA, typically 2k transitions or targets (entries in inclusion list) are automatically defines based on the processing of the LC-OzID-MS DIA dataset. Transitions in Skyline within automated processing steps can reach 500000 for comprehensive analysis of natural samples in full discovery mode.

Public repository transitions:
None, all transitions are independent of repositories

Transitions for different collision energies:
None, CID is not the fragmentation technique employed

Peak detection/alignment:
A combination of Skyline and python processing is used to ensure peak detection and alignment. Quantification of peaks can be manually verified and adjusted in Skyline while running the workflow.

Spectra annotation:
Algorithms for several data analysis steps are written in python and published on the github repository

Quantification:
Validated against small subset of 37 analytical standards as well as NIST 1950 SRM original sample and literature values. Quantification for newly discovered fatty acids is a sophisticated best estimate.

Interactive analysis/visualization:
The analysis is interactive and employs Skyline and Excel files for visualization support during data processing.

Section 4 — Lipid Identification (Untargeted)

Section 4.1 — Full MS (HRAM LC-MS)

Section 4.2 — Data Dependent Acquisition (DDA)

Section 4.3 — Data Independent Acquisition (DIA)

Section 5 — Lipid Quantification (Untargeted)

Section 6 — Analysis and Visualisation