A batch correction method for liquid chromatography–mass spectrometry data that does not depend on quality control samples

TY - JOUR

T1 - A batch correction method for liquid chromatography–mass spectrometry data that does not depend on quality control samples

AU - Rusilowicz, Martin James

AU - Dickenson, Michael

AU - Charlton, Adrian

AU - O'Keefe, Simon

AU - Wilson, Julie C.

N1 - © Authors 2016. This content is made available by the publisher under a Creative Commons Attribution Licence. This means that a user may copy, distribute and display the resource providing that they give credit. Users must adhere to the terms of the licence.

PY - 2016/2/18

Y1 - 2016/2/18

N2 - The need for reproducible and comparableresults is of increasing importance in non-targeted metabolomicstudies, especially when differences betweenexperimental groups are small. Liquid chromatography–mass spectrometry spectra are often acquired batch-wise sothat necessary calibrations and cleaning of the instrumentcan take place. However this may introduce further sourcesof variation, such as differences in the conditions underwhich the acquisition of individual batches is performed.Quality control (QC) samples are frequently employed as ameans of both judging and correcting this variation. Herewe show that the use of QC samples can lead to problems.The non-linearity of the response can result in substantialdifferences between the recorded intensities of the QCs andexperimental samples, making the required adjustmentdifficult to predict. Furthermore, changes in the responseprofile between one QC interspersion and the next cannotbe accounted for and QC based correction can actuallyexacerbate the problems by introducing artificial differences.‘‘Background correction’’ methods utilise allexperimental samples to estimate the variation over timerather than relying on the QC samples alone. We comparenon-QC correction methods with standard QC correctionand demonstrate their success in reducing differences between replicate samples and their potential to highlightdifferences between experimental groups previously hiddenby instrumental variation.

AB - The need for reproducible and comparableresults is of increasing importance in non-targeted metabolomicstudies, especially when differences betweenexperimental groups are small. Liquid chromatography–mass spectrometry spectra are often acquired batch-wise sothat necessary calibrations and cleaning of the instrumentcan take place. However this may introduce further sourcesof variation, such as differences in the conditions underwhich the acquisition of individual batches is performed.Quality control (QC) samples are frequently employed as ameans of both judging and correcting this variation. Herewe show that the use of QC samples can lead to problems.The non-linearity of the response can result in substantialdifferences between the recorded intensities of the QCs andexperimental samples, making the required adjustmentdifficult to predict. Furthermore, changes in the responseprofile between one QC interspersion and the next cannotbe accounted for and QC based correction can actuallyexacerbate the problems by introducing artificial differences.‘‘Background correction’’ methods utilise allexperimental samples to estimate the variation over timerather than relying on the QC samples alone. We comparenon-QC correction methods with standard QC correctionand demonstrate their success in reducing differences between replicate samples and their potential to highlightdifferences between experimental groups previously hiddenby instrumental variation.

U2 - 10.1007/s11306-016-0972-2

DO - 10.1007/s11306-016-0972-2

M3 - Article

SN - 1573-3890

VL - 12

JO - Metabolomics

JF - Metabolomics

IS - 56

ER -