LICRE: Unsupervised feature correlation reduction for lipidomics

Gerard Wong, Jeffrey Chan, Bronwyn A. Kingwell, Christopher Leckie, Peter J. Meikle

Research output: Contribution to journalArticleResearchpeer-review

Abstract

Motivation: Recent advances in high-throughput lipid profiling by liquid chromatography electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) have made it possible to quantify hundreds of individual molecular lipid species (e.g. fatty acyls, glycerolipids, glycerophospholipids, sphingolipids) in a single experimental run for hundreds of samples. This enables the lipidome of large cohorts of subjects to be profiled to identify lipid biomarkers significantly associated with disease risk, progression and treatment response. Clinically, these lipid biomarkers can be used to construct classification models for the purpose of disease screening or diagnosis. However, the inclusion of a large number of highly correlated biomarkers within a model may reduce classification performance, unnecessarily inflate associated costs of a diagnosis or a screen and reduce the feasibility of clinical translation. An unsupervised feature reduction approach can reduce feature redundancy in lipidomic biomarkers by limiting the number of highly correlated lipids while retaining informative features to achieve good classification performance for various clinical outcomes. Good predictive models based on a reduced number of biomarkers are also more cost effective and feasible from a clinical translation perspective. Results: The application of LICRE to various lipidomic datasets in diabetes and cardiovascular disease demonstrated superior discrimination in terms of the area under the receiver operator characteristic curve while using fewer lipid markers when predicting various clinical outcomes.

Original languageEnglish
Pages (from-to)2832-2833
Number of pages2
JournalBioinformatics
Volume30
Issue number19
DOIs
Publication statusPublished - 2 Apr 2014
Externally publishedYes

Cite this

Wong, Gerard ; Chan, Jeffrey ; Kingwell, Bronwyn A. ; Leckie, Christopher ; Meikle, Peter J. / LICRE : Unsupervised feature correlation reduction for lipidomics. In: Bioinformatics. 2014 ; Vol. 30, No. 19. pp. 2832-2833.
@article{8d7f2c0d552d4d28bb66fe93aec6c1bf,
title = "LICRE: Unsupervised feature correlation reduction for lipidomics",
abstract = "Motivation: Recent advances in high-throughput lipid profiling by liquid chromatography electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) have made it possible to quantify hundreds of individual molecular lipid species (e.g. fatty acyls, glycerolipids, glycerophospholipids, sphingolipids) in a single experimental run for hundreds of samples. This enables the lipidome of large cohorts of subjects to be profiled to identify lipid biomarkers significantly associated with disease risk, progression and treatment response. Clinically, these lipid biomarkers can be used to construct classification models for the purpose of disease screening or diagnosis. However, the inclusion of a large number of highly correlated biomarkers within a model may reduce classification performance, unnecessarily inflate associated costs of a diagnosis or a screen and reduce the feasibility of clinical translation. An unsupervised feature reduction approach can reduce feature redundancy in lipidomic biomarkers by limiting the number of highly correlated lipids while retaining informative features to achieve good classification performance for various clinical outcomes. Good predictive models based on a reduced number of biomarkers are also more cost effective and feasible from a clinical translation perspective. Results: The application of LICRE to various lipidomic datasets in diabetes and cardiovascular disease demonstrated superior discrimination in terms of the area under the receiver operator characteristic curve while using fewer lipid markers when predicting various clinical outcomes.",
author = "Gerard Wong and Jeffrey Chan and Kingwell, {Bronwyn A.} and Christopher Leckie and Meikle, {Peter J.}",
year = "2014",
month = "4",
day = "2",
doi = "10.1093/bioinformatics/btu381",
language = "English",
volume = "30",
pages = "2832--2833",
journal = "Bioinformatics",
issn = "1367-4803",
publisher = "Oxford University Press, USA",
number = "19",

}

LICRE : Unsupervised feature correlation reduction for lipidomics. / Wong, Gerard; Chan, Jeffrey; Kingwell, Bronwyn A.; Leckie, Christopher; Meikle, Peter J.

In: Bioinformatics, Vol. 30, No. 19, 02.04.2014, p. 2832-2833.

Research output: Contribution to journalArticleResearchpeer-review

TY - JOUR

T1 - LICRE

T2 - Unsupervised feature correlation reduction for lipidomics

AU - Wong, Gerard

AU - Chan, Jeffrey

AU - Kingwell, Bronwyn A.

AU - Leckie, Christopher

AU - Meikle, Peter J.

PY - 2014/4/2

Y1 - 2014/4/2

N2 - Motivation: Recent advances in high-throughput lipid profiling by liquid chromatography electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) have made it possible to quantify hundreds of individual molecular lipid species (e.g. fatty acyls, glycerolipids, glycerophospholipids, sphingolipids) in a single experimental run for hundreds of samples. This enables the lipidome of large cohorts of subjects to be profiled to identify lipid biomarkers significantly associated with disease risk, progression and treatment response. Clinically, these lipid biomarkers can be used to construct classification models for the purpose of disease screening or diagnosis. However, the inclusion of a large number of highly correlated biomarkers within a model may reduce classification performance, unnecessarily inflate associated costs of a diagnosis or a screen and reduce the feasibility of clinical translation. An unsupervised feature reduction approach can reduce feature redundancy in lipidomic biomarkers by limiting the number of highly correlated lipids while retaining informative features to achieve good classification performance for various clinical outcomes. Good predictive models based on a reduced number of biomarkers are also more cost effective and feasible from a clinical translation perspective. Results: The application of LICRE to various lipidomic datasets in diabetes and cardiovascular disease demonstrated superior discrimination in terms of the area under the receiver operator characteristic curve while using fewer lipid markers when predicting various clinical outcomes.

AB - Motivation: Recent advances in high-throughput lipid profiling by liquid chromatography electrospray ionization tandem mass spectrometry (LC-ESI-MS/MS) have made it possible to quantify hundreds of individual molecular lipid species (e.g. fatty acyls, glycerolipids, glycerophospholipids, sphingolipids) in a single experimental run for hundreds of samples. This enables the lipidome of large cohorts of subjects to be profiled to identify lipid biomarkers significantly associated with disease risk, progression and treatment response. Clinically, these lipid biomarkers can be used to construct classification models for the purpose of disease screening or diagnosis. However, the inclusion of a large number of highly correlated biomarkers within a model may reduce classification performance, unnecessarily inflate associated costs of a diagnosis or a screen and reduce the feasibility of clinical translation. An unsupervised feature reduction approach can reduce feature redundancy in lipidomic biomarkers by limiting the number of highly correlated lipids while retaining informative features to achieve good classification performance for various clinical outcomes. Good predictive models based on a reduced number of biomarkers are also more cost effective and feasible from a clinical translation perspective. Results: The application of LICRE to various lipidomic datasets in diabetes and cardiovascular disease demonstrated superior discrimination in terms of the area under the receiver operator characteristic curve while using fewer lipid markers when predicting various clinical outcomes.

UR - http://www.scopus.com/inward/record.url?scp=84911429285&partnerID=8YFLogxK

U2 - 10.1093/bioinformatics/btu381

DO - 10.1093/bioinformatics/btu381

M3 - Article

VL - 30

SP - 2832

EP - 2833

JO - Bioinformatics

JF - Bioinformatics

SN - 1367-4803

IS - 19

ER -