Format

Send to

Choose Destination
Metabolomics. 2018 Sep 20;14(10):128. doi: 10.1007/s11306-018-1420-2.

Characterization of missing values in untargeted MS-based metabolomics data and evaluation of missing data handling strategies.

Author information

1
Institute of Computational Biology, Helmholtz-Zentrum München, Neuherberg, Germany.
2
Institute of Epidemiology II, German Research Center for Environmental Health, Helmholtz Zentrum München, Neuherberg, Germany.
3
Research Unit of Molecular Epidemiology, German Research Center for Environmental Health, Helmholtz Zentrum München, Neuherberg, Germany.
4
German Center for Diabetes Research (DZD e.V.), Neuherberg, Germany.
5
Institute of Bioinformatics and Systems Biology, Helmholtz-Zentrum München, Neuherberg, Germany.
6
Institute of Experimental Genetics, Genome Analysis Center, Helmholtz Zentrum München, Neuherberg, Germany.
7
Lehrstuhl für Experimentelle Genetik, Technische Universität München, Freising, Germany.
8
German Center for Cardiovascular Disease Research (DZHK e.V.), Munich, Germany.
9
Department of Physiology and Biophysics, Weill Cornell Medical College in Qatar, Education City, Doha, Qatar.
10
Institute of Genetic Epidemiology, Helmholtz Zentrum München-German Research Center for Environmental Health, Neuherberg, Germany.
11
Chair of Genetic Epidemiology, Institute of Medical Informatics, Biometry and Epidemiology, Ludwig-Maximilians-University, Munich, Germany.
12
MRC Epidemiology Unit, University of Cambridge, Cambridge, UK.
13
Department of Mathematics, Technische Universität München, Garching, Germany.
14
German Center for Diabetes Research (DZD e.V.), Neuherberg, Germany. g.kastenmueller@helmholtz-muenchen.de.
15
Institute of Bioinformatics and Systems Biology, Helmholtz-Zentrum München, Neuherberg, Germany. g.kastenmueller@helmholtz-muenchen.de.
16
Institute of Computational Biology, Helmholtz-Zentrum München, Neuherberg, Germany. jan.krumsiek@helmholtz-muenchen.de.
17
German Center for Diabetes Research (DZD e.V.), Neuherberg, Germany. jan.krumsiek@helmholtz-muenchen.de.
18
Institute for Computational Biomedicine, Englander Institute for Precision Medicine, Department of Physiology and Biophysics, Weill Cornell Medicine, New York, USA. jan.krumsiek@helmholtz-muenchen.de.

Abstract

BACKGROUND:

Untargeted mass spectrometry (MS)-based metabolomics data often contain missing values that reduce statistical power and can introduce bias in biomedical studies. However, a systematic assessment of the various sources of missing values and strategies to handle these data has received little attention. Missing data can occur systematically, e.g. from run day-dependent effects due to limits of detection (LOD); or it can be random as, for instance, a consequence of sample preparation.

METHODS:

We investigated patterns of missing data in an MS-based metabolomics experiment of serum samples from the German KORA F4 cohort (n = 1750). We then evaluated 31 imputation methods in a simulation framework and biologically validated the results by applying all imputation approaches to real metabolomics data. We examined the ability of each method to reconstruct biochemical pathways from data-driven correlation networks, and the ability of the method to increase statistical power while preserving the strength of established metabolic quantitative trait loci.

RESULTS:

Run day-dependent LOD-based missing data accounts for most missing values in the metabolomics dataset. Although multiple imputation by chained equations performed well in many scenarios, it is computationally and statistically challenging. K-nearest neighbors (KNN) imputation on observations with variable pre-selection showed robust performance across all evaluation schemes and is computationally more tractable.

CONCLUSION:

Missing data in untargeted MS-based metabolomics data occur for various reasons. Based on our results, we recommend that KNN-based imputation is performed on observations with variable pre-selection since it showed robust results in all evaluation schemes.

KEYWORDS:

Batch effects; K-nearest neighbor; Limit of detection; MICE; Mass spectrometry; Missing values imputation; Untargeted metabolomics

Supplemental Content

Full text links

Icon for PubMed Central
Loading ...
Support Center