LipidFinder

Advanced pipeline for lipidomics discovery applications

Description

LipidFinder is an open-source Python workflow which searches a number of different databases to obtain putative identification of lipids, and assigns them to a class based on the LIPID MAPS classification system. The software quickly distinguishes and quantifies lipid-like features from contaminants, adducts, in-source fragments, isotopes, salt clusters and noise in high resolution liquid chromatography/mass spectrometry (LC/MS) datasets that have been pre-aligned using XCMS or any other preprocessing tool provided the required input file format is preserved. Bear in mind the approach is not suitable for shotgun lipidomics, MS/MS or low resolution datasets.

Technical Information

Publications:

PMID:33027502,

PMID:30101336

Training datasets:

Yes

Documentation and user guide:

Yes

Download / Web-service link:

Github ,

LipidMaps

Source code repository:

Yes

Programming languages:

Python

Platforms:

Windows,

Linux,

MacOS

Output formats:

CSV,

Excel,

PDF

Input formats:

CSV,

TSV,

Excel

Web platform:

Yes

Desktop client:

Yes

CLI:

Yes

GUI:

Yes

License:

MIT

Tasks

4.1) Full MS (HRAM LC-MS)

Other features:

Volcano plot analysis,

Orthogonal partial least-squares discriminant analysis,

Random-forest analysis and ANOVA (only on web-service version)

Batch processing:

Yes

Lipid quantification:

Yes

Lipid identification:

Built-in matching to COMP_DB (Computationally generated database composed of bulk species), LMSD (Complete LIPID MAPS structure database), and curated subset of LMSD.

Correction for in source fragments:

Yes

Support multiple adducts:

Yes

Adjustable feature detection:

Yes

Rt alignment and correction:

Yes

Main workflow steps:

Background removal, feature detection, deconvolution.

LC-MS and LC-MS/MS:

LC-MS, high-resolution MS data are required; XCMS exported data