LipidBlast in silico tandem mass spectrometry database for lipid identification

Abstract

Current tandem mass spectral libraries for lipid annotations in metabolomics are limited in size and diversity. We provide a freely available computer-generated tandem mass spectral library of 212,516 spectra covering 119,200 compounds from 26 lipid compound classes, including phospholipids, glycerolipids, bacterial lipoglycans and plant glycolipids. We show platform independence by using tandem mass spectra from 40 different mass spectrometer types including low-resolution and high-resolution instruments.

Figure 1. Creation, validation and application of in-silico generated tandem mass spectra in LipidBlast

(a) New lipid compound structures were created using combinatorial chemistry approaches. A scaffold of the lipid core structure and linker are connected to fatty acyls with different chain lengths and different degrees of unsaturation. (b) The reference tandem spectra (upper panel) are used to simulate the mass spectral fragmentations and ion abundances of the in-silico spectra (lower panel). The compound shown here is phosphatidylcholine PC(16:0/16:1) at precursor m/z=732.55 [M+H]⁺. (c) Tandem mass spectra are obtained from LC-MS/MS or direct-infusion experiments. The MS/MS spectra are submitted to LipidBlast MS/MS search. An m/z precursor ion filter serves as first powerful filter and a subsequent product ion match creates a library hit score that is related to the level of confidence for the compound annotation.

Figure 2. LipidBlast was mostly developed with ion trap tandem mass spectra but can be used with data from other platforms such as QTOF mass spectrometers

a) The Cardiolipin example shows that even in the in the case of the non-matching but abundant precursor ion at m/z 1239.8355 [M-H]⁻, the correct result is obtained with LipidBlast. b) The standard reference compound with precursor m/z=793.4841 [M-H]⁻ is correctly identified as phosphatidylinositol PI(17:0/14:1) as first hit in a library search with LipidBlast.