Lipid Data Analyzer (LDA)

Extendable chromatographic-based DDA data processing tool for untargeted mass spectrometry supporting the detection of novel lipid species

Description

LDA provides high‑throughput accurate lipid quantification and identification for LC-MS data, as well as specific structural annotation of lipid species using MSn spectra acquired in data-dependent mode. Specificity for MS based identification is increased by automatically removing isotopic peaks, and by taking the isotopic intensity pattern into account. MS/MS based identification relies on decision rule sets, which are a reflection of an MSn interpretation process by a trained expert. LDA is MS platform independent and easily extendable to additional lipid subclasses and adducts. For quantification, the high accuracy of chromatographic peak integration is achieved by an algorithm that detects peak borders in both the m/z and retention time dimensions. Depending on the resolution of the MS platform, this algorithm can differentiate closely eluting isobaric species and even isotopologues in tracing experiments.

Technical Information

Publications:

PMID 34600364,

PMID:29058722,

PMID:33003696,

PMID:32578539,

PMID:21169379

Training datasets:

Yes

Documentation and user guide:

Yes

Download / Web-service link:

Yes

Source code repository:

Yes

Programming languages:

Java

Platforms:

Windows,

Linux,

MacOS

Output formats:

TSV,

Excel,

mzTab-M,

rdb

Input formats:

mzML,

mzXML,

.d(Agilent),

.d(Bruker),

.wiff(SCIEX),

.wiff.scan(SCIEX),

.raw(Thermo),

.raw(Waters)

Desktop client:

Yes

CLI:

Yes

GUI:

Yes

License:

GPL (academic and commercial)

Web platform:

Tasks

4.1) Full MS (HRAM LC-MS)

Other features:

3D chromatogram visualization.,

For RP chromatography, a class‑specific filter is available for removal of features at unlikely retention time by retention time prediction based on number of carbon atoms and double bonds; quantitation can be carried out in parallel, and runs on CUDA compatible GPUs

Batch processing:

Yes

Lipid identification:

Matching to customizable mass lists in Excel (Example mass lists for common lipid classes are provided)

Correction for in source fragments:

Support multiple adducts:

Yes

Algorithms:

Peak border localization in m/z and retention time dimensions, followed by an isotopic intensity pattern check. Matching to customizable mass lists in Excel (example mass lists for common lipid classes are provided)

LC-MS and LC-MS/MS:

LC-MS and LC-MS/MS, PRM

Shotgun MS/MS:

4.2) Data dependent acquisition (DDA)

Other features:

3D spectra visualization

Batch processing:

Yes

Correction for in source fragments:

Support multiple adducts:

Yes

Spectra annotation:

Yes

Scores:

No scoring: unambiguous assignments are indicated in green, and ambiguous assignments (which may be co-eluting isobaric/isomeric species of other subclasses) are indicated in yellow or orange

Algorithms:

MS1 accuracy and isotopic pattern check, MSn rule-based assignment

Requirements:

Customizable Excel mass lists and rule sets consisting of rules for the definition of fragments, their intensity relations, and rules for differentiating between potential isomeric/isobaric candidates; rule sets can be used for lipid class verification (head group), FA/LCB chain annotation and sn-position assignment

Decision rules:

Yes

Spectra matching:

LC-MS and LC-MS/MS:

Yes

Shotgun MS/MS:

4.5) Identification of oxidized lipids

Other features:

Validation of ambiguous hits using closest theoretical m/z

Spectra annotation:

Yes

Scores:

No scoring: validation through rule sets and closest theoretical m/z

Algorithms:

On the fly mass calculation

Requirements:

Oxidation levels defined in mass list (provided and customizable)

Decision rules:

Yes

Spectra matching:

Lipid coverage:

oxo-, keto-, epoxy-, furan-, hydroxyl-, epidioxy-, carboxy-, hydroperoxy-, cyclopentane-, bromo-, chloro-, fluoro-, iodo-, and nitro- for 4 out of 8 LIPID MAPS categories

5) Lipid quantification from untargeted lipidomics datasets (HRAM MS, DDA, DIA)

Calibration curve:

Semi absolute quantification using internal standards:

Yes

Relative quantification:

Yes

Normalization:

Yes

Quantified using:

Peak volume.

Adjustable feature detection:

Smoothing in time and m/z dimension; border detection; delimitation of overlapping signals.