U.S. Department of Energy

Pacific Northwest National Laboratory

Breaking a Bottleneck in LC-MS/MS Lipidomics Studies

LIQUID interface. (A) File input. (B) Search parameters. (C) Table of output. Checked boxes denote validated identifications. The highlighted species, PI (18:1/20:4), is selected as an example for the remaining displayed features. (D) MS/MS spectra with annotated diagnostic (red) and fragment ions (green). (E) List of observed MS/MS fragments, including their structural annotations. The theoretical MS/MS fragments for the associated lipid species are also available. (F) Isotopic profile of the lipid species selected for MS/MS analysis and generated using data from MS scans. (G) Extracted ion chromatogram (XIC) for the precursor ion. The peak apex is highlighted in red while the nearest MS level scan is highlighted green. Peak start and stop scan numbers can be input for determination of peak area, which is shown in (H). (H) Target lipid information.

The Science

There is a key bottleneck in lipidomics studies based on liquid chromatography coupled with tandem mass spectrometry (LC-MS/MS): Researchers are unable to process spectral data in a high-throughput manner and at the same time confidently identify and accurately quantify the lipid species detected.

Incorrect identifications can result in misleading biological interpretations. Yet existing tools are not designed for high-volume verification of identifications and have to be manually verified to ensure accuracy. Since larger scale studies are increasingly desired, analysts need improved software for identifying lipids.

A recent paper by led author Jennifer E. Kyle and eight co-authors at Pacific Northwest National Laboratory (PNNL) introduces an open-source lipid identification software, Lipid Quantification and Identification (LIQUID). The scoring is trainable, the search database is customizable, and multiple lines of evidence are displayed (allowing for confident identification to be made). Single- and global-target as well as fragment-pattern searches are also available so that similar and repeating patterns of MS/MS spectra can be tracked.

The Impact

LIQUID enables high-throughput lipidomics data analysis and for the semi-quantitative identification of lipid species from LC-MS/MS data.


LIQUID enables users to process high-throughput lipidomics LC-MS/MS data. The software contains a customizable target library and a scoring model for different project needs.

The graphics user interface allows visualization of multiple lines of spectral evidence for each lipid identification. This lets researchers rapidly examine data for confident identifications of lipid molecular species. The software also enables users to specify the number of identification results per MS/MS scan, which allows for the identification of co-eluting species.

Confidently identified lipids areexported in an output file. Once exported, those files can be loaded later if re-analysis of the data is needed.

Compared to other freely available software commonly used to identify lipids and other small molecules, LIQUID has a rapid processing time that can generate a higher number of validated lipididentifications faster. Its reference database includes more than 21,200 unique lipid targets across six lipid categories, 24 classes, and 63 subclasses. LIQUID is able to confidently identify more lipid species with a faster combined processing and validation time than any other software in its field. 

What’s Next?

What is next for LIQUID is to increase the reference library to include lipids that may be unique to particular disease states or to organisms from select environmental niches.  This will enable greater characterization of the diverse range of samples received at PNNL and therefore, enhance our understanding of biological and environmental systems of interest.  


This work was supported by the National Institute of Allergy and Infectious Diseases, and by the Department of Energy Office of Biological and Environmental Research, Genomic Science Program under the Pacific Northwest National Laboratory Pan-omics project. 


Kyle JE, Crowell KL, Casey CP, Fujimoto GM, Kim S, Dautel SE, Smith RD, Payne SH, Metz TO. “LIQUID: an-open source software for identifying lipids in LC-MS/MS-based lipidomics data.” Bioinformatics. (2017). Jun 1;33(11):1744-1746. DOI: 10.1093/bioinformatics/btx046.

June 2017
| Pacific Northwest National Laboratory