Aarhus University Seal

Dansk

You are here: » Assessment of XCMS Optimization Methods with Machine-Lear...

About Aarhus University

Assessment of XCMS Optimization Methods with Machine-Learning Performance

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

Standard

Assessment of XCMS Optimization Methods with Machine-Learning Performance. / Lassen, Johan; Nielsen, Kirstine Lykke; Johannsen, Mogens et al.

In: Analytical Chemistry, Vol. 93, No. 40, 10.2021, p. 13459-13466.

Research output: Contribution to journal/Conference contribution in journal/Contribution to newspaper › Journal article › Research › peer-review

Harvard

Lassen, J, Nielsen, KL, Johannsen, M & Villesen, P 2021, 'Assessment of XCMS Optimization Methods with Machine-Learning Performance', Analytical Chemistry, vol. 93, no. 40, pp. 13459-13466. https://doi.org/10.1021/acs.analchem.1c02000

APA

Lassen, J., Nielsen, K. L., Johannsen, M., & Villesen, P. (2021). Assessment of XCMS Optimization Methods with Machine-Learning Performance. Analytical Chemistry, 93(40), 13459-13466. https://doi.org/10.1021/acs.analchem.1c02000

CBE

Lassen J, Nielsen KL, Johannsen M, Villesen P. 2021. Assessment of XCMS Optimization Methods with Machine-Learning Performance. Analytical Chemistry. 93(40):13459-13466. https://doi.org/10.1021/acs.analchem.1c02000

MLA

Lassen, Johan et al. "Assessment of XCMS Optimization Methods with Machine-Learning Performance". Analytical Chemistry. 2021, 93(40). 13459-13466. https://doi.org/10.1021/acs.analchem.1c02000

Vancouver

Lassen J, Nielsen KL, Johannsen M, Villesen P. Assessment of XCMS Optimization Methods with Machine-Learning Performance. Analytical Chemistry. 2021 Oct;93(40):13459-13466. doi: 10.1021/acs.analchem.1c02000

Author

Lassen, Johan ; Nielsen, Kirstine Lykke ; Johannsen, Mogens et al. / Assessment of XCMS Optimization Methods with Machine-Learning Performance. In: Analytical Chemistry. 2021 ; Vol. 93, No. 40. pp. 13459-13466.

Bibtex

@article{20da8d662f9140d093749df153f824ce,
title = "Assessment of XCMS Optimization Methods with Machine-Learning Performance",
abstract = "The metabolomics field is under rapid development. In particular, biomarker identification and pathway analysis are growing, as untargeted metabolomics is usable for discovery research. Frequently, new processing and statistical strategies are proposed to accommodate the increasing demand for robust and standardized data. One such algorithm is XCMS, which processes raw data into integrated peaks. Multiple studies have tried to assess the effect of optimizing XCMS parameters, but it is challenging to quantify the quality of the XCMS output. In this study, we investigate the effect of two automated optimization tools (Autotuner and isotopologue parameter optimization (IPO)) using the prediction power of machine learning as a proxy for the quality of the data set. We show that optimized parameters outperform default XCMS settings and that manually chosen parameters by liquid chromatography-mass spectrometry (LC-MS) experts remain the best. Finally, the machine-learning approach of quality assessment is proposed for future evaluations of newly developed optimization methods because its performance directly measures the retained signal upon preprocessing.",
author = "Johan Lassen and Nielsen, {Kirstine Lykke} and Mogens Johannsen and Palle Villesen",
note = "Publisher Copyright: {\textcopyright} 2021 American Chemical Society",
year = "2021",
month = oct,
doi = "10.1021/acs.analchem.1c02000",
language = "English",
volume = "93",
pages = "13459--13466",
journal = "Analytical Chemistry",
issn = "0003-2700",
publisher = "AMER CHEMICAL SOC",
number = "40",

}

RIS

TY - JOUR

T1 - Assessment of XCMS Optimization Methods with Machine-Learning Performance

AU - Lassen, Johan

AU - Nielsen, Kirstine Lykke

AU - Johannsen, Mogens

AU - Villesen, Palle

N1 - Publisher Copyright: © 2021 American Chemical Society

PY - 2021/10

Y1 - 2021/10

N2 - The metabolomics field is under rapid development. In particular, biomarker identification and pathway analysis are growing, as untargeted metabolomics is usable for discovery research. Frequently, new processing and statistical strategies are proposed to accommodate the increasing demand for robust and standardized data. One such algorithm is XCMS, which processes raw data into integrated peaks. Multiple studies have tried to assess the effect of optimizing XCMS parameters, but it is challenging to quantify the quality of the XCMS output. In this study, we investigate the effect of two automated optimization tools (Autotuner and isotopologue parameter optimization (IPO)) using the prediction power of machine learning as a proxy for the quality of the data set. We show that optimized parameters outperform default XCMS settings and that manually chosen parameters by liquid chromatography-mass spectrometry (LC-MS) experts remain the best. Finally, the machine-learning approach of quality assessment is proposed for future evaluations of newly developed optimization methods because its performance directly measures the retained signal upon preprocessing.

AB - The metabolomics field is under rapid development. In particular, biomarker identification and pathway analysis are growing, as untargeted metabolomics is usable for discovery research. Frequently, new processing and statistical strategies are proposed to accommodate the increasing demand for robust and standardized data. One such algorithm is XCMS, which processes raw data into integrated peaks. Multiple studies have tried to assess the effect of optimizing XCMS parameters, but it is challenging to quantify the quality of the XCMS output. In this study, we investigate the effect of two automated optimization tools (Autotuner and isotopologue parameter optimization (IPO)) using the prediction power of machine learning as a proxy for the quality of the data set. We show that optimized parameters outperform default XCMS settings and that manually chosen parameters by liquid chromatography-mass spectrometry (LC-MS) experts remain the best. Finally, the machine-learning approach of quality assessment is proposed for future evaluations of newly developed optimization methods because its performance directly measures the retained signal upon preprocessing.

UR - http://www.scopus.com/inward/record.url?scp=85117206136&partnerID=8YFLogxK

U2 - 10.1021/acs.analchem.1c02000

DO - 10.1021/acs.analchem.1c02000

M3 - Journal article

C2 - 34585906

AN - SCOPUS:85117206136

VL - 93

SP - 13459

EP - 13466

JO - Analytical Chemistry

JF - Analytical Chemistry

SN - 0003-2700

IS - 40

ER -