Greazy: Open-Source Software for Automated Phospholipid Tandem Mass Spectrometry Identification.

Fri, 2016-06-24 11:57 -- voskuhlt
TitleGreazy: Open-Source Software for Automated Phospholipid Tandem Mass Spectrometry Identification.
Publication TypeJournal Article
Year of Publication2016
AuthorsKochen MA, Chambers MC, Holman JD, Nesvizhskii AI, Weintraub ST, Belisle JT, M Islam N, Griss J, Tabb DL
JournalAnal Chem
Volume88
Issue11
Pagination5733-41
Date Published2016 Jun 7
ISSN1520-6882
Abstract

Lipid identification from data produced with high-throughput technologies is essential to the elucidation of the roles played by lipids in cellular function and disease. Software tools for identifying lipids from tandem mass (MS/MS) spectra have been developed, but they are often costly or lack the sophistication of their proteomics counterparts. We have developed Greazy, an open source tool for the automated identification of phospholipids from MS/MS spectra, that utilizes methods similar to those developed for proteomics. From user-supplied parameters, Greazy builds a phospholipid search space and associated theoretical MS/MS spectra. Experimental spectra are scored against search space lipids with similar precursor masses using a peak score based on the hypergeometric distribution and an intensity score utilizing the percentage of total ion intensity residing in matching peaks. The LipidLama component filters the results via mixture modeling and density estimation. We assess Greazy's performance against the NIST 2014 metabolomics library, observing high accuracy in a search of multiple lipid classes. We compare Greazy/LipidLama against the commercial lipid identification software LipidSearch and show that the two platforms differ considerably in the sets of identified spectra while showing good agreement on those spectra identified by both. Lastly, we demonstrate the utility of Greazy/LipidLama with different instruments. We searched data from replicates of alveolar type 2 epithelial cells obtained with an Orbitrap and from human serum replicates generated on a quadrupole-time-of-flight (Q-TOF). These findings substantiate the application of proteomics derived methods to the identification of lipids. The software is available from the ProteoWizard repository: http://tiny.cc/bumbershoot-vc12-bin64 .

DOI10.1021/acs.analchem.6b00021
Alternate JournalAnal. Chem.
PubMed ID27186799
Grant ListR01 GM107148 / GM / NIGMS NIH HHS / United States
T15 LM007450 / LM / NLM NIH HHS / United States
U01 AI115619 / AI / NIAID NIH HHS / United States
U01 HL122642 / HL / NHLBI NIH HHS / United States
U24 DK097153 / DK / NIDDK NIH HHS / United States