Electrophoresis 2018, 39, 909–923 909

Biswapriya B. Misra

Department of Internal Medicine,Section of Molecular Medicine,Medical Center Boulevard,Winston-Salem, NC, USA

Received November 18, 2017Revised December 17, 2017Accepted December 18, 2017

Review

New tools and resources in metabolomics:2016–2017

Rapid advances in mass spectrometry (MS) and nuclear magnetic resonance (NMR)-basedplatforms for metabolomics have led to an upsurge of data every single year. Newer high-throughput platforms, hyphenated technologies, miniaturization, and tool kits in data ac-quisition efforts in metabolomics have led to additional challenges in metabolomics datapre-processing, analysis, interpretation, and integration. Thanks to the informatics, statis-tics, and computational community, new resources continue to develop for metabolomicsresearchers. The purpose of this review is to provide a summary of the metabolomicstools, software, and databases that were developed or improved during 2016–2017, thus,enabling readers, developers, and researchers access to a succinct but thorough list ofresources for further improvisation, implementation, and application in due course oftime.

Keywords:

Data / Network / Software / Statistics / Visualization / WorkflowDOI 10.1002/elps.201700441

1 Introduction

Metabolomics, the study of small molecules with a molecu-lar weight of 50–2000 Da (i.e., metabolites), is an emergingfield which studies biological processes. More specifically,metabolomics studies cells, tissues, organs, and biofluidsfrom organisms or environmental samples. This relativelynew field of –omics has led to advances in the understandingof human, animal, plant, microbial, and environmental com-position and functions. As a tool for hypothesis generationand as a means for understanding metabolism and physiol-ogy from a systems biology perspective, metabolomics hasbeen widely used in recent studies across multiple fields.In addition to efforts in other –omics which are closer togenotype such as genomics, epigenomics, and transcrip-tomics (RNAomics), proteomics and metabolomics advanceshave led us closer to the phenotype that can help decipher

Correspondence: Dr. Biswapriya B. Misra, Department ofInternal Medicine, Section of Molecular Medicine, Medical CenterBoulevard, Winston-Salem, NC 27157, USAE-mail: bbmisraccb@gmail.com

Abbreviations: CIL, Chemical isotope labeling; DB,Database; GUI, Graphical User Interface; HRMS,High-resolution mass spectrometry; KEGG, Kyoto En-cyclopedia of Genes and Genomes; MSI, Imagingmass spectrometry; PCA, Principal component analysis;PLS-DA, Partial least square discriminant analysis; QCs,Quality control sample; QqQ, Triple quadruple; Q-ToF, Hybridquadrupole orthogonal time-of-flight; R, R-Statistical pro-gramming language; UPLC-TOF, Ultra performance liquidchromatography time-of-flight mass spectrometry; XCMS,Various forms (X) of chromatography mass spectrometry

organismal behavior and functions, define health and dis-ease statuses, and provide translatable evidence for agricul-ture, therapeutics, pharmaceuticals, environmental sciences,and personalized (precision) medicine efforts. Metabolomicsas a research tool has touched all aspects of healthcare andmedicine. Some of the most recent studies in 2017 thathave used metabolomics have included studies on pain [1],reproductive medicine [2], human breath [3], sweat analy-sis [4], coronary atherosclerosis [5], autophagy regulation [6],system-wide coverage from single cells to entire planet [7],plant metabolomics [8], and exposome scale metabolomics [9]among others.

There is a continual increase in sophistication, highthroughputness, and in advancements of the analytical plat-forms used for the generation of metabolomics datasets- bothfor nuclear magnetic resonance (NMR) and mass spectrom-etry (MS)-based platforms (and associated improvements insample preparation, chromatography, etc.). However, thesenewer platforms have posed heightened challenges in han-dling, pre-processing, statistical treatment, visualization, andinterpretation of datasets. Thankfully, to this end, a highly-engaged community of informatics, statistics, computational,and data scientists are well equipped in developing and im-proving on the existing and latest tools, databases, and re-sources for the metabolomics-research dependent investi-gators. A recently conducted survey highlighted the trendsand opportunities for tool development in the Galaxy work-flow based on the different platforms and data-types [10].Advances in the next generation of platforms and technol-ogy for capillary electrophoresis mass spectrometry (CE-MS)

Color Online: See the article online to view Figs. 1 and 2 in color.

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

910 B. B. Misra Electrophoresis 2018, 39, 909–923

-based metabolomics looks promising as well [11]. Groupsof researchers have also provided a very detailed set ofprotocols for metabolomics and lipidomics using traveling-wave ion mobility mass spectrometry (IMS) [12]. Very ex-tensive and detailed reviews have captured the future ofmetabolomics research [13] and as avenue for global chemicalanalysis [14].

Very recently, Spicer et al. [15] provided a list of freely-available open source software tools for metabolomics analy-sis. In addition, they created a GitHub resource containing alist which is classified and searchable via a simple controlledvocabulary [16]. Newer databases (DBs) such as the MINEDB was used for epimetabolite annotation [17]. In anothereffort, almost 50 DBs were reviewed, from which 17 wereselected for the online electron spray ionization mass spec-trometry (ESI-MS) functionality of the DBs [18] where the au-thors underscored the general characteristics and functionsof each DB from their experience derived at the Centre forMetabolomics and Bioanalysis (CEMBIO) in Madrid.

Advancements in analytical and theoretical methods arecontinually updated in online resources, such as catalogsof metabolomic tools such as OMICtools (http://omic-tools.com/) [19], Fiehn Lab resources (http://fiehnlab.ucda-vis.edu/ and http://metabolomics.ucdavis.edu/Downloads),the (International) Metabolomics Society’s resource pages(http://metabolomicssociety.org/resources/metabolomics-databases), metabomatch (http://www.metabomatch.com/softwares/) catalog, and software repositories such asComprehensive R Archive Network (CRAN) (https://cran.r-project.org/web/packages/available_packages_by_name.html), Bioconductor (https://www.bioconductor.org/) andGitHub (https://github.com/). In addition to cataloging thecommon tools, a summary of their highlighted featuresand major uses is also valuable. Building on the previouslyestablished review structure [20, 21], this overview of majortools and resources in metabolomics is organized into thefollowing sections: (i) Gas chromatography mass spec-trometry (GC-MS) tools, (ii) Liquid chromatography massspectrometry (LC-MS) tools, (iii) Imaging mass spectrometry(MSI) tools, (iv) Nuclear magnetic resonance (NMR) tools,(v) Untargeted data analysis tools, (vi) Ion mobility MS tools,(vii) Tools for isotope techniques, (viii) Data pre-processingtools, (ix) Annotation tools, (x) Multifunctional tools, and (xi)Miscellaneous tools of interest. Table 1 shows a summary ofall the reviewed resources in this article.

2 Tools for analytical platforms

2.1 Gas chromatography mass spectrometry

(GC-MS) tools

Gas chromatography mass spectrometry (GC-MS) remainsa very handy yet robust tool for routine metabolomics re-search for most research groups and core facilities. Addition-ally, GC-MS also complements liquid chromatography massspectrometry (LC-MS) based approaches, by virtue of its more

efficient capability of capturing the volatile and polar chem-ical space of biological materials. With the advent of newerplatforms such as high resolution (HR) GC-MS in the formof proprietary GC-OrbiTrap technology [22–24], efforts arestill very active for development of comprehensive softwaretools. For instance, new DB-building and MS subtraction al-gorithms have been developed for automated, sequential two-dimensional gas chromatography/mass spectrometry (GC xGC-MS) [25]. These algorithms have allowed the first use ofa DB building tool, with full mass spectrum subtraction, thatdoes not rely on high resolution MS data. A comprehensivelist of tools widely-used in 2017 is provided in the followingsection.

Mixture model normalization (mixnorm) accommodatestruncated GC-MS- based metabolomics data and estimatesper-metabolite batch and run-order effects using quality con-trol (QC) samples [26]. Mixnorm outperforms other ap-proaches across many metrics, including improved correla-tion of non-targeted and targeted measurements and superiorperformance when metabolite detectability varies accordingto batch.

Baitmet allows high-throughput and wide scope inter-rogation on the metabolic composition of complex samplesanalyzed using GC–MS via freely available spectral data [27].

MetaboloDerivatizer was developed to calculate the chem-ical properties of derivatized compounds and all retentionindexes in EI–MS DBs are predicted using a simple mathe-matical model [28] (Fig. 1A). Furthermore, the MS-FINDERspectral search engine was developed and utilized for search-ing EI–MS DBs using mass spectral similarity with the eval-uation of false discovery rate (FDR).

PARAFAC2 based Deconvolution and Identification System(PARADISe) is a standalone tool and approach for processingraw GC-MS data for fast conversion of raw CDF files into peaktables using minimum steps [29].

M-IOLITE is a computational suite for the efficient andautomatic analysis of high-throughput metabolomics datafrom GC-MS [30] with an excellent graphical user interface(GUI).

R2DGC is an R-package for two-dimensional GC-MS dataanalysis that implements retention time (RT) and MS simi-larity threshold-free alignments, performs ion filtering, andis compatible with multiple peak quantification methods [31].

ADAP-GC is an R-based automated computational work-flow for extracting metabolite information from raw, untar-geted gas GC-MS-metabolomics data [32].

MIDcor is an R-program that performs a primary analysisof isotopic isomers (isotopomers) distribution obtained fromGC-MS data [33].

2.2 Liquid chromatography mass spectrometry

(LC-MS) tools

Liquid chromatography mass spectrometry (LC-MS)-basedmetabolomics approaches remain the cornerstone ofmetabolomics data generation for routine and sophisticatedapplications by virtue of advancements in chromatography

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

Electrophoresis 2018, 39, 909–923 General 911

Table 1. The following list of reviewed tools is organized by major areas of metabolomics and includes details regarding implementatione.g. programming language (R, C/C++, MATLAB, Python, Java, etc.) or web browser based and usability. Each tool’s ease of useis specified as difficult (needing in-depth understanding and knowledge of programming language applicable for expert users),medium (modest level of proficiency or programming skills applicable for intermediate users) and easy (requires minor level ofskill to implement the tool and applicable for beginners)

Tools Implementation Ease of use Software availability References

GC-MSBaitmet R Difficult https://cran.r-project.org/web/packages/baitmet/index.html [27]MetaboloDerivatizer Standalone Easy http://prime.psc.riken.jp/Metabolomics Software/ [28]PARADISe Standalone Easy http://www.models.life.ku.dk/paradise [29]M-IOLITE NA Easy http://miolite.iceht.forth.gr:8080/ [30]R2DGC R Difficult https://github.com/HudsonAlpha/R2DGC [31]ADAP-GC 3.2 R Difficult https://github.com/du-lab/ADAP-GC-3.0 [32]MIDcor R Difficult https://github.com/seliv55/mid correct [33]LC-MSMass Studio Visual Studio

2013 R©Easy NA [34]

MAIMS Python Difficult https://github.com/savantas/MAIMS [35]MS-FLO Web Easy http://msflo.fiehnlab.ucdavis.edu/#/ [36]MetaboQC R Difficult https://cran.r-project.org/web/packages/MetaboQC/index.html [38]LIQUID R Difficult https://github.com/PNNL-Comp-Mass-Spec/LIQUID [39]AntDAS MATLAB Difficult http://software.tobaccodb.org/software/antdas [40]FPIC Python Difficult https://github.com/zmzhang/pymass [41]KPIC2 R Difficult https://github.com/hcji/KPIC2 [42]Lipostar NA Difficult http://download.moldiscovery.com/lipostar-1.0.0b12-win64.exe [43]PiMP Web Easy http://polyomics.mvls.gla.ac.uk/ [96]Imaging MSpySM, Python Difficult https://github.com/alexandrovteam/pySM [44]massPix R Difficult https://github.com/hallz/massPix [45]NMRNMRProcFlow Web/

StandaloneEasy http://nmrprocflow.org/ [46]

NMRSpec MATLAB Difficult https://hub.docker.com/r/nmrprocflow/nmrspec/ [47]MWASTools R Difficult https://bioconductor.org/packages/MWASTools/ [49]speaq 2.0 R Difficult https://github.com/beirnaert/speaq [50]PepsNMR R Difficult https://github.com/ManonMartin/PepsNMR [51]Plasmodesma Python Difficult https://github.com/delsuc/plasmodesma [52]GISSMO Web Easy http://gissmo.nmrfam.wisc.edu/ [53]UntargetedMetaNetter 2.0 Java Easy Cytoscape App Store [55]metaX R, Web Easy http://metax.genomics.cn/ [56]MetCirc R Difficult https://github.com/PlantDefenseMetabolism/MetCirc [60]Ion MobilityMetCCS Web Easy http://www.metabolomics-shanghai.org/MetCCS/ [61]LipidCCS Web Easy http://www.metabolomics-shanghai.org/LipidCCS/ [62]PIXiE R Difficult https://github.com/PNNL-Comp-Mass-Spec/PIXiE

https://github.com/PNNL-Comp-Mass-Spec/IMS-Informed-Library[63]

Isotope-AnalysisMetExtract II R Medium https://metabolomics-ifa.boku.ac.at/metextractII/ [65]AssayR R Difficult https://gitlab.com/jimiwills/assay.R [66]Preprocessing/QCMetImp Web Easy https://metabolomics.cc.hawaii.edu/software/MetImp/ [68]Gsimp R Difficult https://github.com/WandeRum/GSimp [69]NOREVA Web Easy http://server.idrb.cqu.edu.cn/noreva/ [70]B-MIS R Difficult https://github.com/IngallsLabUW/B-MIS-normalization [71]MetaboDrift Microsoft

Excel,Macros

Easy https://github.com/evanscr/MetaboDrift [72]

SERRF R Difficult http://www.serrf.fiehnlab.ucdavis.edu/ocpu/library/SERRFweb/www/ [74]sampleDrift R Difficult https://gitlab.com/CarlBrunius/sampleDrift [75]

(Continued)

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

912 B. B. Misra Electrophoresis 2018, 39, 909–923

Table 1. Continued

Tools Implementation Ease of use Software availability References

AnnotationxMSannotatot R Easy https://sourceforge.net/projects/xmsannotator/ [37]compMS2Miner R, Web Easy https://github.com/WMBEdmands/compMS2Miner [58]UC2 Java-based

GUIDifficult http://webs2.kazusa.or.jp/mfsearcher [59]

creDBle DB Web Easy http://credble.wustl.edu/ [80]Imet Web Easy http://imet.seeslab.net [82]ms2lda Web Easy http://www.ms2lda.org/ [83]Curatr R, Web Easy http://curatr.mcf.embl.de/ [86]LipidHunter Python Difficult https://bitbucket.org/SysMedOs/lipidhunter [89]CEU Mass Mediator Web Easy http://ceumass.eps.uspceu.es/index.xhtml [90]LDA Java-based

GUIEasy http://genome.tugraz.at/lda2/lda description.shtml [91]

LPPtiger software Python Difficult https://bitbucket.org/SysMedOs/lpptiger [92]LipidCircos Python Difficult https://bitbucket.org/SysMedOs/lipidcircos [92]GlycoPAT MATLAB Difficult https://www.virtualglycome.org/glycopat [122]MultifunctionalproFIA R Difficult http://bioconductor.org/packages/proFIA [73]Workflow4Metabolomics

3.0Galaxy, Web Easy http://workflow4metabolomics.org/ [93]

MetaboAnalystR R Difficult https://github.com/xia-lab/MetaboAnalystR [94]MetaBox Web Easy http://kwanjeeraw.github.io/metabox/ [95]OthersMIDAS-G C++ Difficult http://facultyweb.mga.edu/yingfeng.wang/Assets/midasg/midas-g.htm [57]msPurity R Difficult https://bioconductor.org/packages/release/bioc/html/msPurity.html [79]mixOmics R Difficult http://mixomics.org/ [88]mzAccess Web Medium http://www.mzaccess.org/ [101]MASTR-MS Web Difficult https://muccg.github.io/mastr-ms/ [104]MetabolitePredict Web Easy http://xulab.case.edu/MetabolitePredict [105]MetExploreViz Web Easy http://metexplore.toulouse.inra.fr/metexploreViz/doc/ [106]NA DH Web Easy http://nadh.ice.mpg.de/NaDH/ [107]OmicsDI DB Easy http://www.nature.com/nbt/journal/v35/n5/fig tab/nbt.3790 F1.html [108]YMDB 2.0 DB Easy http://www.ymdb.ca/ [109]ODG Java Difficult https://github.com/jguhlin/odg [110]FlavorDB Web Easy http://cosylab.iiitd.edu.in/flavordb [111]PAMDB Web Easy http://pseudomonas.umaryland.edu/ [112]MetaComp C, R Easy http://cqb.pku.edu.cn/ZhuLab/MetaComp/ [113]EDA Web Easy https://eda.nc3rs.org.uk/ [114]visJS2jupyter Python Difficult https://pypi.python.org/pypi/visJS2jupyter [116]MAGI Web Easy https://magi.nersc.gov/ [118]zeroSum R Difficult https://github.com/rehbergT/zeroSum [119]RankProd 2.0 R Difficult https://www.bioconductor.org/packages/devel/bioc/html/RankProd.html [120]RNAModMapper R Difficult http://bearcatms.uc.edu/new/limbachgroup publication/rna-modmapper-

software.html[121]

Metadata Analyser R Difficult https://github.com/lasigeBioTM/MetadataAnalyser [123]SATORI Web Easy http://satori.refinery-platform.org/ [124]BIOCONDA Multiple Medium https://bioconda.github.io/ [125]GRITS Toolbox Java Easy http://www.grits-toolbox.org/ [126]PySCeSToolbox Python Difficult https://github.com/PySCeS/PyscesToolbox [127]ChemRICH Web/R Easy http://chemrich.fiehnlab.ucdavis.edu/ocpu/library/ChemRICHTest3/www/ [128]

as well as high resolution (HR) tandem mass spectrometry(MS/MS). Tools and software made available to the analyti-cal chemistry and metabolomics community are introducedbelow.

MASS Studio is a software utility that simplifies LC-MS traces to allow metabolomics analysis on large sets of

microbial natural products samples in a drug discoveryproject environment [34].

MAIMS is an open-source Python (Python 2.7 and3.5 + compliant) script for command line usage and isan extendible tool for isotopologue profile deconvolutiontasks [35].

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

Electrophoresis 2018, 39, 909–923 General 913

Figure 1. Snapshots of the tools and resources described in this article. (A) MetaboloDerivatizer showing the silylation prediction ofL-cysteine, as an example, from input SMILES, (B) NMRProcFlow interface showing options and workflow, (C) GISSMO DB showing theexperimental and theoretical spectrum overlays for L-cysteine as an example, and (D) MetaboDrift interface on Microsoft Excel Macros.

Mass Spectral Feature List Optimizer (MS-FLO) is a webapplication to improve the quality of feature lists after initialprocessing to expedite data curation [36]. The tool utilizes RTalignments, accurate mass tolerances, Pearson’s correlationanalysis, and peak height similarity to identify ion adducts,duplicate peak reports, and isotopic features of the mainmonoisotopic metabolites.

xMSannotator, implemented in R, is an algorithm thatuses HR-MS data for a series of samples with commonproperties and publicly available chemical, metabolic, andenvironmental DBs to assign confidence levels to annota-tion results [37]. The MS/MS evaluation of a set of randomlyselected 210 metabolites, annotated using xMSannotator inan untargeted metabolomics data set from human, showsthat 80% of features with high or medium confidence scoreshave ion dissociation patterns consistent with xMSannota-tor‘s findings. The annotation results are obtained by query-ing local or online DBs such as ChemSpider, Kyoto Encyclo-pedia of Genes and Genomes (KEGG), Human MetabolomeDatabase (HMDB), Toxin and Toxin Target Database (T3DB),and LipidMaps.

MetaboQC is an R package which helps to correct intra-day and inter-days variability using QCs analyzed within apre-set sequence of experiments [38]. MetaboQC was testedin two data sets to assess the correction effects by compar-ing the metabolites variability before and after application of

the proposed tool where the number of entities in QCs wassignificantly different between days was reduced from 86 to19% in the negative ionization mode and from 100 to 13% inthe positive ionization mode.

LIQUID is a tool with GUI for semi-automated process-ing and visualization of LC-MS/MS-based lipidomics data[39]. The tool allows users to process high-throughput dataand access a customizable target library and scoring modelper project needs which includes visualization of multiplelines of spectral evidence for each lipid identified. When com-pared to other open-source tools used in lipid and metabo-lite identification, (e.g. Competitive Fragmentation Modelingidentification (CFM-ID), MetFrag, Global Natural ProductSocial Molecular Networking (GNPS), LipidBlast and MS-DIAL), the tool demonstrated faster processing time and ahigher number of matches.

AntDAS (Automatic Data Analysis Strategy for UPLC-QTOF-based Nontargeted Metabolic Profiling Analysis) is aMATLAB tool with GUI that allows automated extraction ofchromatographic peaks in the UPLC–QTOF data, a peak an-notation stage which clusters fragment ions that belong tothe same metabolite, a time-shift correction across the sam-ples, registration of components by using an adaptive net-work searching algorithm, a list of statistical methods, andcompound identification by matching against a compoundlibrary with � 500 plant metabolites [40].

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

914 B. B. Misra Electrophoresis 2018, 39, 909–923

Fast Pure Ions Chromatograms (FPIC) is a Python im-plemented package for extracting pure ion chromatograms(PICs) from raw LC-MS datasets effectively and quickly [41].

KPIC2 is an R-package for extracting meaningful PICswhich allows for their alignment, grouping for identificationof isotope and adducts, and filling missing peaks and multi-variate pattern recognition among others [42].

Lipostar is a software tool for targeted and untargeted LC-MS lipidomics data analysis for data acquisition, multivariateanalysis, and lipid identification with or without the supportof lipid DBs [43].

2.3 Imaging mass spectrometry (MSI) tools

Imaging mass spectrometry (MSI) is rapidly gaining popu-larity for providing visual and localization information formetabolite distribution ranging from plant cells to humantissues. Towards this end, the tools and resources are contin-uing to add up for the research community and those reportedin 2017 are described here.

For HR- imaging mass spectrometry in particular, pySMis a Python-based framework for FDR-controlled metaboliteannotation at the level of the molecular sum formula [44].The tool introduces a metabolite-signal match score and atarget–decoy FDR estimate for spatial metabolomics as well.

EM is another tool that produces single ion images, per-forms multivariate statistics and provides putative lipid anno-tations based on accurate mass matching against generatedlipid libraries [45].

2.4 Nuclear magnetic resonance (NMR) tools

From ease of sample preparation, to robustness in quantita-tion, NMR-based metabolomics continues to deliver qualitymetabolomics data that allows more accurate metabolite iden-tification and quantification. The tools and resources madeavailable to the metabolomics community this year are intro-duced below.

NMRProcFlow is dedicated to metabolic fingerprintingand targeted metabolomics that covers all spectra processingsteps including baseline correction, chemical shift calibra-tion, and alignment [46] (Fig. 1B).

NMRSpec, is another tool for pre-processing and ana-lyzing NMR data, which consists of four major functionalblocks: spectrum alignment, baseline correction, peaks ex-traction, and peak integration [47].

nmrML is an open XML-based exchange and storage for-mat for NMR spectral data [48]. The nmrML format is in-tended to be fully compatible with existing NMR data forchemical, biochemical, and metabolomics experiments. Itcaptures raw NMR data, spectral data acquisition parame-ters, and spectral metadata such as chemical structures. Ad-ditionally, it is compatible with pure-compound NMR datafor reference spectral libraries and NMR data from complexbio-mixtures from metabolomics experiments.

MWASTools is an R package designed to provide anintegrated pipeline to analyze metabonomic data in large-scale epidemiological studies. It allows for QC analysis,metabolome-wide association analysis using a set of mod-els, statistical visualization, and metabolite assignmentusing statistical total correlation spectroscopy (STOCSY)workflows, and interpretation of metabolome-wide associa-tion study (MWAS) results [49].

speaq 2.0 is an R package that eases NMR-based dataanalysis and offers spectral processing such as raw spectralalignment and processing as well as peak based processingsteps [50]. Moreover, it facilitates analysis of complementaryLC-MS workflows.

PepsNMR is an R package for pre-processing of 1H-NMRbased metabolomics data [51]. The tool performs solvent sig-nal suppression, internal calibration, phase, baseline and mis-alignment corrections, bucketing and normalization. It alsoallows for faster workflows, reproducibility, and flexible re-porting among other features.

Plasmodesma is a Python-based package that performsautonomous, unsupervised processing of a large corpus of1D, 2D, and Diffusion Ordered SpectroscopY (DOSY) ex-periments from a series of samples acquired in differentconditions [52].

GISSMO (Guided Ideographic Spin System Model Opti-mization) is a GUI-based webserver for guided ideographicspin system model optimization, which enables the ef-ficient calculation and refinement of spin system matri-ces against experimental 1D-1H NMR spectra of smallmolecules [53] (Fig. 1C).

3 Tools for data analysis

3.1 Untargeted data analysis tools

Untargeted metabolomics approaches aid in discovery ofnovel biomarkers and in the process of unknown discoveryand identification. A recent review summarized the untar-geted LC-MS data preprocessing in metabolomics and pop-ular tools such as XCMS, MZmine, MAVEN, and Metabo-Analyst among other popular tools [54]. Tools for untargetedmetabolomics have always kept the community interested,and recently highlighted methods are introduced below.

MetaNetter 2 is a Cytoscape plugin for ab initio networkanalysis and metabolite feature classification [55]. It allowsboth chemical transformation and adduct mapping, whereRT windows are supported for both adduct and transformnetwork generation.

metaX is an R package with web interface for untargetedmetabolomics data analysis that relies on a set of interchange-able modules that provides functionalities such as peak pick-ing and annotation, data quality assessment, missing valueimputation, data normalization, univariate and multivariatestatistics, power analysis, and sample size estimation. Addi-tional features include receiver operating characteristic (ROC)

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

Electrophoresis 2018, 39, 909–923 General 915

analysis, biomarker selection, pathway annotation, correla-tion network analysis, and metabolite identification [56].

MIDAS ((Metabolite Identification via DB Searching)-G is a C++ implementation of the MIDAS algorithmfor investigating fragmentation rules where it processes ametabolite as a simple graph and uses graph grammarsto recognize specific chemical bonds and their adjacentstructures [57].

compMS2Miner (Comprehensive MS2 Miner) is an Rpackage that aids in rapid, comprehensive feature annota-tion using a peak-picker-output and MS2 data files as in-puts [58]. Indeed, CompMS2Miner integrates many usefultools in a single workflow for metabolite annotation andprovides an overview of the MS2 data with a web applica-tion GUI compMS2Explorer (Comprehensive MS2 Explorer)for data sharing and reproducibility.

UC2 (Unique Connectivity of Uncharged Compounds) isa search system where compounds are tentatively neutralizedinto uncharged states and stored on the basis of their uniqueconnectivity of atoms after removing their stereo chemicalinformation using the first block in the hash of the IUPACInternational Chemical Identifier, by which false-positive hitsare remarkably reduced, both charged and uncharged com-pounds are properly searched in a single query, and recordshaving a unique connectivity are compiled in a single searchresult [59].

MetCirc is a Bioconductor package that allows alignmentand calculation of pairwise similarity scores among MS/MSdata and visualization of these across samples [60]. Basedon the compound familial groupings, it allows discovery ofshared metabolites and hypothesis formulation for unknowndiscovery.

3.2 Ion mobility MS tools

The expanding field of metabolomics has seen ion mobilityMS being applied to numerous studies and the future lookspromising for this new platform. Predicting the collisionalcross sections (CCS) values towards assigning identities ofmetabolites remains critical for the advancement of the plat-form towards analysis of complex samples. To this end, thedeveloped tools are introduced below.

MetCCS Predictor is a web-server that allows prediction ofCCS values of metabolites, in ion mobility MS experimentsby using molecular descriptors within a few seconds; thusenabling metabolite identification [61].

LipidCCS (lipid collision cross-section) is a web-server thatallows prediction of CCS values of lipids in ion mobility MSexperiments using molecular descriptors within a few sec-onds, thus, enabling metabolite identification [62]. The preci-sion in prediction in LipidCCS DB (15 646 lipids and 63 434CCS values in total) can potentially reduce false-positive iden-tifications of lipids.

PNNL Ion Mobility Cross Section Extractor (PIXiE) is anR package which allows automatic extraction of the arrivaltimes for molecules measured by drift tube ion mobilityspectrometry coupled with mass spectrometry (DTIMS-MS)

at multiple electric fields and compute their associated CCS[63].

3.3 Tools for isotope techniques

In vivo and in vitro labeling with stable isotopes (SIL) of-fers a wide array of advantages in performing quantita-tive MS based metabolomics experiments. In addition, thecorrection of natural isotope in MS/MS measurements inmetabolomics and 13C fluxomics experiments remain criti-cal [64]. Therefore, tools aiding in identification and quan-tification of labelled metabolites, screening of isotopologues,in deisotoping of spectra, etc. are useful resources for thecommunity. Tools developed towards this end are describedbelow.

MetExtract II is a software for detection of biologicallyderived compounds which exploits SIL-specific isotope pat-terns and elution profiles in LC-HRMS(/MS) data [65]. It con-sists of three complementary modules: M1 (AllExtract), M2(TracExtract), and M3 (FragExtract) aiding in identification ofall labelled ions.

AssayR is an R package that takes HR wide-scan LC-MS data sets and tailors peak detection for each metabo-lite through a simple, iterative user interface by automati-cally integrating peak areas for all isotopologues and outputsextracted ion chromatograms (EICs), absolute and relativestacked bar charts for all isotopologues, and a .csv data file [66].

AccuCor is an algorithm that aids in natural isotope cor-rection for HR MS labeling data which takes into accountwhether isotopologues with the same nominal mass areresolved [67].

3.4 Data pre-processing tools

Typical data preprocessing tools deal with noise filtering,baseline correction, peak detection, deisotoping, alignment,accurate mass calculation, identification, and normalizationamong others. In 2017, several data pre-processing tools werereported as these have been areas of very active research.These tools define the quality of data generated from a spe-cific platform for further interpretation and some of the mostwidely used are discussed below.

MetImp combined “modified 80% rule” and imple-mented a comprehensive strategy for missing value imputa-tion in metabolomics data and developed a public-accessibleweb-tool [68] (Fig. 2A).

GSimp is an iterative Gibbs sampler based, left-censoredmissing value imputation approach that is implemented asan R package for imputing missing values in targeted missingvalues that are missing not at random (MNAR) [69].

NOREVA (Normalization and Evaluation of MS-basedmetabolomics data) is a web server for helping remove un-wanted signal variation by implementing the most completeset of the available normalization methods. It has uniquefeatures that remove overall unwanted variations based onquality control (QC) metabolites and allows for QC

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

916 B. B. Misra Electrophoresis 2018, 39, 909–923

Figure 2. Snapshots of the tools and resources described in this article. (A)MetImp interface showing multiple options for imputingmetabolomics data in a very user friendly manner, (B) WorkFlow4Metabolomics workflow showing the wide array of options it offers, (C)Nicotiana DB showing metabolite-metabolite co-expression map, (D) Flavor DB showing circular map of sources of flavors and chemicals,(E) ChemRICH output showing tree-map of related metabolites and an enrichment view. (F) OmicsDI interface showing the wide rangeof DBs it allows data to archive from.

sample-based correction sequentially followed by data nor-malization [70].

B-MIS (best-matched internal standard) normalizationmethod, implemented in R, is applicable for targeted and un-targeted metabolomics datasets and yields relative concentra-tion, using the power of matching the measured metaboliteswith isotope-labeled internal standards that behave similarlyduring the analysis [71].

MetaboDrift, implemented in Macro-enabled MicrosoftExcel, is a tool for visualization and correction of driftin multi-batch LC-MS metabolomics data where driftcorrection is comparable to isotope standard normalization-based quantification [72] (Fig. 1D).

proFIA is an R package that implements several novelalgorithms to preprocess Flow Injection Analysis coupled toHigh-Resolution Mass Spectrometry (FIA-HRMS) workflows

for high throughput metabolomics studies [73]. The tool per-forms noise estimation, peak detection and quantification,peak grouping across samples, and missing value imputa-tion among other functionalities.

SERRF (Systematic Error Removal using Random For-est) is a QC-based normalization method for large-scale un-targeted metabolomics, which uses the machine-learning al-gorithm Random Forest to normalize the data [74].

SampleDrift is an R package that explores the effects ofpre-centrifugation storage on plasma sample quality by al-lowing predictive modeling (storage temperature and time)and kinetic modelling as well as drift correction for NMR-data [75] most commonly, but can be extendible to LC-MS datasets.

In a recent review, the statistical methods and workflowfor analyzing human metabolomics data were laid out [76] to

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

Electrophoresis 2018, 39, 909–923 General 917

assist researchers in choosing the most relevant and precisetools. The authors delineated the approaches in several stepssuch as study design, data management, optional simulationanalyses, cross-sectional or prospective analyses, visualiza-tion of findings, and prioritization of results for follow-upinvestigations among others.

3.5 Annotation tools

Mass spectrometry annotation tools are typically partitionedby the level of annotation provided into: (a) unequivocalmolecular formula, (b) tentative candidates and (c) libraryspectrum match [77]. Recently, towards identification of smallmolecules using accurate mass (HR) MS/MS search, severalDBs were extensively discussed elsewhere [78]. Listed beloware the 2017 tools that were either not considered in the afore-mentioned review or were very recently released.

msPurity is a vendor-independent R package for LC-MS and direct infusion (DI) MS2 that calculates a simplemetric to describe the contribution of the selected precur-sor [79]. The package was applied to data dependent acquisi-tion (DDA)-based MS2 metabolomics data sets derived fromthree metabolomics data repositories and can be used to as-sess precursor purity of theoretical isolation windows fromLC-MS data sets of differing sample types.

creDBle DB is an alternative approach to untargetedmetabolomics that relies on thoroughly annotated referencedata sets, where the DB contains accurate mass, RT, andMS/MS fragmentation data as well as annotations of all cre-dentialed features [80].

Sirius is a standalone tool for fragmentation pattern anal-ysis and fragmentation tree computation from MS/MS data[81].

iMet is a tool that facilitates structural annotation ofmetabolites not described in DBs [82]. iMet uses MS/MSspectra and the exact mass of an unknown metabolite toidentify metabolites in a reference DB that are structurallysimilar to the unknown metabolite.

MS2LDA is a method for the decomposition of sets ofmolecular fragment data derived from large metabolomicsexperiments [83]. ms2lda.org is a web application that allowsusers to upload their data, run MS2LDA analyses, and explorethe results through interactive visualizations.

Direct Infusion Metabolite DB (DIME DB) is a DB ofbiologically relevant metabolite structures and annotationswith 95,043 freely accessible metabolites [84]. The DB offersopen access to thousands of biologically relevant metaboliteannotations, sourced from many respected metabolite DBs.Whereas other metabolite DBs limit the user to rudimentaryweb browser-based search facilities, DIMEdb offers a pow-erful Representational state transfer application program in-terface (RESTful API). This API provides programmatic end-points from which tools can be built to interoperate with theDB and ease the metabolite identification process.

MassBank of North America (MoNA) is an open accessmass spectral DB for small molecules, with a corpus of

spectral fragmentation data from ESI and electron impactionization (EI) sources [85].

Curatr is a web application for rapid generation of highquality mass fragmentation libraries for LC-MS analysis andit supports the export of the spectral libraries in standardformats for downstream applications [86].

BioCAn (Biologically Consistent Annotation), combinesthe results from DB searches and in silico fragmentation anal-yses and places these results into a relevant biological contextfor the sample as captured by a metabolic model [87].

mixOmics is an R package for -omics feature selectionand multiple data integration such as transcriptomics, pro-teomics, and metabolomics [88]. The package allows data ex-ploration, dimension reduction and visualization for adoptinga systems biology approach in –omics data analysis.

LipidHunter is a Python-based package for high through-put identification of phospholipids acquired via LC-MS andshotgun experimental workflows, where identification isbased on MS/MS data analysis that follows defined fragmen-tation roles for each phospholipid class [89]. Furthermore,the identified signals are tested for elemental compositionand bulk identification via the LIPID MAPS search with out-puts such as information-rich tabular and graphical reports.

CMM (CEU Mass Mediator) is an annotation webserverthat supports queries from MS based experimental mass, overmultiple metabolomic data sources such as KEGG, Metlin,LipidMaps and HMDB, followed by filtering and scoring theannotations based on rules that represent expert knowledge[90].

LDA (Lipid Data Analyzer) is a standalone tool that al-lows automated and reliable annotation of lipids and theirmolecular structures from HR XC-MS coupled MS/MS datasets [91].

LPPtiger is a Python-based tool with a GUI for ox-idized phospholipids identification from data-dependentLC-MS datasets [92]. LipidCircos is aimed to generate circosplot for lipidomics data.

3.6 Multifunctional tools

In addition to the platform specific tools or the tools ad-dressing one component of data analysis, there are severalmulti-functional tools that work pretty much in a platform-independent manner, i.e., for CE, LC- and GC-MS and/orNMR generated data sets. Furthermore, these tools performdata preprocessing, statistical analysis, data visualization and,at times, pathway and network analysis. The multi-functionaltools that surfaced this year are described below.

Workflow4Metabolomics 3.0 is one of the multifunctionaltools, which aids in holistic handling and analysis of LC–MS,FIA–MS, GC–MS, and NMR data analysis workflows [93].

MetaboAnalystR contains the R functions and librariesunderlying the popular MetaboAnalyst web server, including500 functions for data processing, normalization, statisticalanalysis, metabolite set enrichment analysis, metabolic path-way analysis, and biomarker analysis [94] (Fig. 2B).

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

918 B. B. Misra Electrophoresis 2018, 39, 909–923

Metabox is an R-based web application for data process-ing, statistical analysis, integrative visual exploration, andfunctional analysis with several approaches (such as func-tional class scoring, overrepresentation analysis, and WordCloud generation) [95].

Polyomics integrated Metabolomics Pipeline (PiMP) offersautomated and user-friendly analysis from MS data acqui-sition to biological interpretation metabolomics data analy-sis [96]. In addition, a ‘Summary Page’ provides a simpleoverview of the experiment in the format of a scientific pa-per, depicting the key findings of the experiment along withassociated metadata. It also includes a ‘Metabolite Page’ thatlists metabolites with ‘evidence cards’, featuring criteria be-hind metabolite annotation (i.e., peak shapes, intensities indifferent sample groups, and DB information).

4 Miscellaneous tools of interest

This section enlists and discusses the features of tools, ap-plication, and packages which are indirectly associated witha metabolomics workflow, but that help in upstream datahandling or downstream data interpretation and analysis, orintegration with other –omics data sets and at times in im-proving the analytical workflow. For instance, tool harmo-nization is also being improved by efforts to containerizetools, such as those by the PhenoMeNal consortium (Phe-noMeNal) and the BioContainers initiative [97]. The effortsand resources made public in 2017 are listed and describedbelow.

WikiPathways (wikipathways.org), a comprehensivepathway DB, was updated recently to allow bridging themetabolomics datasets with other omics research [98]. Con-tinued growth, development, and curation, efforts have re-sulted in, for instance, doubling of the number of annotatedmetabolite nodes in WikiPathways-thus providing an excel-lent resource for machine readable omics data analysis andvisualization resource for multi-omics frameworks.

HMDB 4.0 is an update with nearly three-fold increasein the number of completely annotated metabolites, four-foldincrease in experimental data made available, and a 60-fold rise in the number of illustrated metabolic pathways,thus underscoring its importance in human metabolomicsstudies [99]. In addition, this update includes huge num-ber of predicted MS/MS and GC-MS reference spectral data,metabolite-SNP (single nucleotide polymorphism), and drug-metabolite interaction data.

mzML2ISA (encompassing mzML and imzML formats)and nmrML2ISA (nmrML format only) are a set of Python-based parsing packages that can automatically generateISA-Tab metadata file stubs from raw XML metabolomicsdata files [100]. Overall, the use of mzML2ISA & nm-rML2ISA reduces the time needed to capture metadatasubstantially.

mzAccess is a software that provides interactive, onlineaccess to primary MS data in real-time via a web service,circumventing the need for bulk data processing [101]. The

tool allows querying of instrument data for spectra, chro-matograms, or two-dimensional mz-RT areas in either profileor centroid modes through a simple, uniform interface thatis independent of vendor or instrument type.

mzR is an R package which provides a unified API to thecommon file formats and parsers available for MS data thatcomes with a wrapper for .mzXML, .mzData, and .mzMLfiles [102]. This package holds promise in metabolomicsdata analysis pipelines towards a better harmonization of up-stream data extraction process.

MSS is a flexible network feature selection frameworkthat combines metabolomics data with the genome-scalemetabolic network where the method adopts a sequentialfeature screening procedure and machine learning-based cri-teria to select important subnetworks and identify the optimalfeature matching simultaneously [103]. The method identifiesseveral subnetworks that are supported by the current litera-ture as well as detects some subnetworks with plausible newfunctional implications.

MASTR-MS is a standalone laboratory information man-agement system (LIMS) solution that can be deployed eitherwithin a single laboratory or used to link workflows across amultisite network [104]. It is comprised of (i) a Node Manage-ment System that can be used to link and manage projectsacross one or multiple collaborating laboratories; (ii) a UserManagement System which defines different user groups andprivileges of users; (iii) a Quote Management System whereclient quotes are managed; (iv) a Project Management Systemin which metadata is stored and all aspects of project man-agement, including experimental setup, sample tracking andinstrument analysis, are defined, and (v) a Data ManagementSystem that allows the automatic capture and storage of rawand processed data from the analytical instruments to theLIMS.

MetabolitePredict is a web-based generalizable diseasemetabolite prediction system that uses disease name or aset of disease-associated genes as the input for query [105].

MetExploreViz is an open source web componentthat provides functionalities for visualization of metabolicnetworks and pathways and allows analysis of omics data ina biochemical context [106].

Nicotiana attenuata Data Hub (Na DH) is an integra-tive web platform for exploring genomic, transcriptomic, andmetabolomic data in wild tobacco (Nicotiana species) [107](Fig. 2C). In addition to the standalone unique metabolomicsdata for this model plant, the genomic and transcriptomicdata integration with metabolite would pave way for betterannotation of unknown metabolites as well as novel biosyn-thetic pathways.

Omics Discovery Index (OmicsDI) allows the discoveringand linking of public omics data sets using it’s web inter-face [108] (Fig. 2F). For instance, OmicsDI allows users tolink their meta-data, raw data files, and accession IDs frommultiple DBs such as Metabolights, and Metabolomics Work-bench for metabolomics, and PRIDE for proteomics to NCBIGenBank for transcriptomics and genomics datasets into oneprofile page.

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

Electrophoresis 2018, 39, 909–923 General 919

YMDB 2.0 is a significantly expanded version of the yeastmetabolome DB [109]. The DB’s newer features would helpmove forward the metabolite annotation and pathway discov-ery efforts in this important model organism.

ODG (Omics DB Generator) is a tool for generating,querying, and the analysis of multi-omics comparative DBs tofacilitate biological understanding [110]. ODG imports defi-nitions and annotations from other sources such as Inter-ProScan, the Gene Ontology, ENZYME, and UniPathwayamong others.

FlavorDB is a web server that comprises of 25 595 flavormolecules representing an array of tastes and odors [111] (Fig.2D). Among these, 2254 molecules are associated with 936natural ingredients belonging to 34 categories.

PAMDB (Pseudomonas aeruginosa Metabolome DB) isa curated DB containing extensive metabolomics data andpathway information from Pseudomonas aeruginosa (referencestrain PAO1) with � 4370 metabolites and 5695 proteins inplace [112].

MetaComp is an easy-to-use integrative tool for compar-ative meta-omics for metagenomics, metatranscriptomics,metaproteomics, and metabolomics data analysis [113]. Thetool is capable of automatically choosing an appropriate two-group sample test based upon the traits of input abundanceprofiles. Furthermore, the tool read files generated by sev-eral upstream programs and performs multivariate statistics,hypothesis testing of two-sample, and regression analysis ofenvironmental factors.

EDA is a web-based application with an integrated web-site, which guides researchers through the process of de-signing animal experiments; the output includes a dia-gram that improves the transparency of the experimentalplan [114].

lumpGEM, genome-scale metabolic networks (GEMs),generates biosynthetic subnetworks composed of reactionsthat can synthesize a target metabolite from a set of definedcore precursor metabolite [115].

visJS2jupyter is a tool designed to embed interactive net-works in Jupyter notebooks to streamline network analysisand to promote reproducible research [116].

ElMaven is an open source MS data processing en-gine that is optimal for isotopomer labeling and globalmetabolomic profiling experiments [117]. It is designedto reduce complexity in metabolomics analysis by us-ing an intuitive interface for exploring the experimentaldata.

MAGI (Metabolite, Annotation, and Gene Integration) isa tool to quickly find and score consensus between metabo-lite identifications and gene annotations which generatesmetabolite-gene associations via biochemical reactions basedon a score between probable metabolite identifications andprobable gene annotations [118].

zeroSum is an R package allows the transfer of metabolicsignatures across labs [119]. zeroSum identifies the samebiomarkers regardless of the scaling method. It can be usedfor small data sets and reduced precision. This tool im-poses the zero-sum constraint combined with the elastic net

regularization. It supports both linear regression (Gaussianresponse) and logistic regression (binomial response).

RankProd 2.0, available at Bioconductor as a part of themzMatch pipeline provides a more principled implementa-tion of the statistics for unpaired datasets, and is based onRank Product (RP), a statistical technique widely used to de-tect differentially expressed features in molecular profilingexperiments such as transcriptomics, metabolomics, and pro-teomics studies [120].

RNAModMapper (RAMM) is tool that enables in the in-terpretation of LC-MS/MS data generated from DNA or RNAsequences by creating a local DB against which collision-induced dissociation (CID) MS/MS data of modified oligonu-cleotides (and modified nucleosides) are compared [121].

GlycoPAT (GlycoProteomics Analysis Toolbox) is a MAT-LAB tool with applications in shotgun glycoproteomicsdata analysis. This program allows analysis of LC-MSn

based high-throughput experimental data for the identi-fication of site-specific N- and O-linked glycosylation ofproteins [122].

Metadata Analyser is a tool that measures metadata qual-ity [123]. When applied to Metabolights, the metabolomicsdata repository results show that the tool successfully imple-ments the proposed measures.

SATORI is web-based and ontology-guided visual explo-ration system for data repositories, which combines powerfulmetadata search with a treemap and a node-link diagram thatvisualize the repository structure [124].

Bioconda is a channel for the conda package managerspecializing in bioinformatics software [125]. Bioconda pro-vides dependency and environment management for anylanguage—Python, R, Ruby, Lua, Scala, Java, JavaScript, C/C++, FORTRAN.

GRITS Toolbox is a software that facilitates glycomicsdata processing and archiving by allowing the user to uploadinstrument MS data files in open file formats (i.e., mzXML,mzML) and annotate them using the annotation module Gly-comics Elucidation and Annotation Tool (GELATO) [126].

PySCeSToolbox is a Python-based, extension to thePython Simulator for Cellular Systems (PySCeS) that in-cludes tools for performing generalized supply–demand anal-ysis, symbolic metabolic control analysis, and for kineticsand thermodynamics measurement of enzyme catalyzed re-actions [127].

ChemRICH (Chemical Similarity Enrichment Analysisfor Metabolomics), is a web server facilitating visualizationsbased on chemical ontologies and structural similarity basedgrouping and enrichment of metabolites form metabolomicsexperiments [128] (Fig. 2E). Furthermore, the toolprovides enrichment statistics that are superior to pathwayenrichments and do not rely on the size of the backgroundDB.

Several newly developed R packages targetingmetabolomics analysis are made available every year atthe CRAN repositories. Due to circumstances, such as theauthors have released the package for the community to useprior to manuscript submission, or they have been developed

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

920 B. B. Misra Electrophoresis 2018, 39, 909–923

but not intended for development of manual or manuscripts,these are not associated with publications.

NormalizeMets is an R package for addressing unwantedvariation in metabolomics datasets due to factors such asbatch effects, matrix effects, and confounding biological vari-ation [129]. This package is a collection of functions designedto implement, assess, and choose a suitable normalizationmethod for a given metabolomics study.

5 Concluding remarks

New tools, DBs, resources, and applications continue toemerge every year; however, it is an enormous task to eval-uate and try each of them before passing on a verdict ontheir usefulness, ease or difficulty of usage, thus, it is on thecommunity to analyze and decide effectiveness in the com-ing years. Informing and educating the upcoming genera-tion of researchers in metabolomics remains another chal-lenge for the community. To this end, efforts to encour-age the next generation of researchers into metabolomicscontinue. One such recent endeavor attempted to introduceNMR-based metabolomics to undergraduate students usingcoffee beans [130]. Additionally, efforts based on community-scale questionnaires have also attempted to summarize theactivities, limitations, and future in quality assurance (QA)and quality control (QC) processes in metabolomics research[131]. Moreover, the community still lacks robust, flexible,cloud-based user-friendly data analysis workflows and plug-gable frameworks which integrate all necessary componentsfrom vendor-dependent MS and spectroscopy platforms todata archiving or providing biological insights. Tools for flux-omics study were not included in this review as well. Theresearch community still needs tools and softwares with min-imal use of IT-expertise, computational resources, and robustresults in near-real-time data processing times. Lack of refer-ence datasets (i.e., ‘gold standards’ as being made availablein proteomics), ways to eliminate or minimize false posi-tives, and standardization of preprocessing steps are othermajor limitations with tools that are currently being devel-oped. Moreover, the community is yet to reuse and repurposethe archived datasets in a reproducible manner. Further, is-sues with scripting-language and version controls limit dataaccessibility and harmonization. The interoperability of cur-rent non-compatible software is among other glaring issuesthat remain to be addressed.

The author sincerely acknowledges the talented developers ofthe metabolomics-oriented tools that intend to carry the field for-ward and for their service to the ever-expanding metabolomicscommunity. The author also apologizes to the developers and cre-ators of numerous tools and resources that could not find a place inthis review due to limitation in space, or due to the sheer overlookand whose work was described within a few sentences only. Theauthor would also like to thank the two kind anonymous review-ers for providing constructive insights, acknowledging the effort

intended for the metabolomics research community, and provid-ing an opportunity to submit a revised version of the manuscript.

The author has declared no conflict of interest.

6 References

[1] Barberini, L., Saba, L., Noto, A., Fattuoni, C., Finco, G.,Saba L., (Eds.), Neuroimaging of Pain, Springer Inter-national Publishing, Cham, pp. 19–33.

[2] Bracewell-Milnes, T., Saso, S., Abdalla, H., Nikolau, D.,Norman-Taylor, J., Johnson, M., Holmes, E., Thum, M.Y., Human Reprod. Update 2017, 1–14.

[3] Couto, M., Barbosa, C., Silva, D., Rudnitskaya, A., Del-gado, L., Moreira, A., Rocha, S. M., Pediatr. Allergy Im-munol. 2017, 28, 452–457.

[4] Delgado-Povedano, M. M., Calderon-Santiago, M., deCastro, M. L., Priego-Capote, F., Talanta 2017, 177,47–65.

[5] Gao, X., Ke, C., Liu, H., Liu, W., Li, K., Yu, B., Sun, M.,Sci. Rep. 2017, 7, 11817.

[6] Stryeck, S., Birner-Gruenberger, R., Madl, T., Microbial.Cell 2017 4, 240.

[7] Petras, D., Jarmusch, A. K., Dorrestein, P. C., Curr. Opin.Chem. Biol. .2017, 36, 24–31.

[8] de Souza, L. P., Naake, T., Tohge, T., Fernie, A. R., Giga-Science 2017, 6, 1–20.

[9] Warth, B., Spangler, S., Fang, M., Johnson, C. H., Fors-berg, E. M., Granados, A., Martin, R. L., Domingo-Almenara, X., Huan, T., Rinehart, D., Montenegro-Burke, J. R., Anal. Chem. 2017, 89, 11505–11513.

[10] Weber, R.J., Lawson, T.N., Salek, R.M., Ebbels, T.M.,Glen, R.C., Goodacre, R., Griffin, J.L., Haug, K., Koul-man, A., Moreno, P., Ralser, M., Metabolomics 2017,13, 12.

[11] Zhang, W., Hankemeier, T., Ramautar, R., Curr. Opin.Biotechnol. 2017, 43, 1–7.

[12] Paglia, G., Astarita, G., Nat. Protoc. 2017, 12, 797–813.

[13] van Rijswijk, M., Beirnaert, C., Caron, C., Cascante,M., Dominguez, V., Dunn, W. B., Ebbels, T. M., Gia-comoni, F., Gonzalez-Beltran, A., Hankemeier, T., Haug,K., F1000Research 2017, 6.

[14] Aksenov, A. A., da Silva, R., Knight, R., Lopes, N. P.,Dorrestein, P. C., Nat. Rev. Chem. 2017, 1, 41570-017.

[15] Spicer, R., Salek, R. M., Moreno, P., Canueto, D., Stein-beck, C., Metabolomics 2017, 13, 106.

[16] List of Metabolomics Tools, https://github.com/RASpicer/MetabolomicsTools 2017.

[17] Lai, Z., Kind, T., Fiehn, O. Anal. Chem. 2017, 89,10171–10180.

[18] Gil de la Fuente, A., Armitage, E. G., Otero, A., Bar-bas, C., Godzien, J., Electrophoresis 2017, 38, 2242–2256.

[19] Henry, V. J., Bandrowski, A. E., Pepin, A. S., Gonzalez,B. J., Desfeux, A., Database 2014, p.bau069.

[20] Misra, B. B., der Hooft, J. J., Electrophoresis 2016, 37,86–110.

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

Electrophoresis 2018, 39, 909–923 General 921

[21] Misra, B. B., Fahrmann, J. F., Grapov, D., Electrophore-sis 2017, 38, 2257–2274.

[22] Peterson, A.C., Balloon, A.J., Westphall, M.S., Coon,J.J., Anal. Chem. 2014, 86, 10044–10051.

[23] Peterson, A. C., Hauschild, J. P., Quarmby, S. T.,Krumwiede, D., Lange, O., Lemke, R. A., Grosse-Coosmann, F., Horning, S., Donohue, T. J., Westphall,M. S., Coon, J. J., Anal. Chem. 2014, 86, 10036–10043.

[24] Peterson, A. C., McAlister, G. C., Quarmby, S. T.,Griep-Raming, J., Coon, J. J., Anal. Chem. 2010, 82,8618–8628.

[25] Robbat, A., Kfoury, N., Baydakov, E., Gankin, Y., J. Chro-matogr. A 2017, 1505, 96–105.

[26] Reisetter, A. C., Muehlbauer, M. J., Bain, J. R., Nodzen-ski, M., Stevens, R. D., Ilkayeva, O., Metzger, B. E., New-gard, C. B., Lowe, W. L., Scholtens, D. M., BMC Bioin-formatics 2017, 18, 84.

[27] Domingo-Almenara, X., Brezmes, J., Venturini, G.,Vivo-Truyols, G., Perera, A., Vinaixa, M., Metabolomics2017, 13, 93.

[28] Matsuo, T., Tsugawa, H., Miyagawa, H., Fukusaki, E.,Anal. Chem. 2017, 89, 6766–6773.

[29] Johnsen, L. G., Skou, P. B., Khakimov, B., Bro, R., J.Chromatogr. A. 2017, 1503, 57–64.

[30] Maga-Nteve, C., Klapa, M. I., IFAC-PapersOnLine 2016,49, 286–288.

[31] Ramaker, R. C., Gordon, E., Cooper, S. J., Bioin-formatics 2017, 179168, https://doi.org/10.1093/bioinformatics/btx825.

[32] Smirnov, A., Jia, W., Walker, D. I., Jones, D. P., Du, X.,J. Proteome Res. 2018, 17, 470–478.

[33] Selivanov, V. A., Benito, A., Miranda, A., Aguilar, E.,Polat, I. H., Centelles, J. J., Jayaraman, A., Lee, P. W.,Marin, S., Cascante, M., BMC Bioinformatics 2017, 18,88.

[34] Martınez, G., Gonzalez-Menendez, V., Martın, J., Reyes,F., Genilloud, O., Tormo, J.R., International Confer-ence on Bioinformatics and Biomedical Engineering,Springer, Cham., pp. 230–244.

[35] Verdegem, D., Moseley, H. N., Vermaelen, W., Sanchez,A. A., Ghesquiere, B., Metabolomics 2017, 13, 123.

[36] DeFelice, B. C., Mehta, S. S., Samra, S., C ajka, T.,Wancewicz, B., Fahrmann, J. F., Fiehn, O., Anal. Chem.2017, 89, 3250–3255.

[37] Uppal, K., Walker, D. I., Jones, D. P., Anal Chem 2017,89, 1063–1067.

[38] Calderon-Santiago, M., Lopez-Bascon, M. A., Peralbo-Molina, A., Priego-Capote, F., Talanta 2017, 174, 29–37.

[39] Kyle, J. E., Crowell, K. L., Casey, C. P., Fujimoto, G. M.,Kim, S., Dautel, S. E., Smith, R. D., Payne, S. H., Metz,T. O., Bioinformatics 2017, 33, 1744–1746.

[40] Fu, H.Y., Guo, X.M., Zhang, Y.M., Song, J.J., Zheng,Q.X., Liu, P.P., Lu, P., Chen, Q. S., Yu, Y. J., She, Y., Anal.Chem. 89, 11083–11090.

[41] Wang, R., Ji, H., Ma, P., Zeng, H., Xu, Y., Zhang, Z. M.,Lu, H. M., Chemom. Intell. Lab. Syst. 2017, 170, 68–74.

[42] Ji, H., Zeng, F., Xu, Y., Lu, H., Zhang, Z., Anal. Chem.2017, 89, 7631–7640.

[43] Goracci, L., Tortorella, S., Tiberi, P., Pellegrino, R. M., DiVeroli, A., Valeri, A., Cruciani, G., Anal. Chem. 2017, 89,6257–6264.

[44] Palmer, A., Phapale, P., Chernyavsky, I., Lavigne, R.,Fay, D., Tarasov, A., Kovalev, V., Fuchser, J., Nikolenko,S., Pineau, C., Becker, M., Nat. Methods 2017, 14, 57–60.

[45] Bond, N.J., Koulman, A., Griffin, J.L., Hall, Z.,Metabolomics 2017, 13, 128.

[46] Jacob, D., Deborde, C., Lefebvre, M., Maucourt, M., Mo-ing, A., Metabolomics 2017, 13, 36.

[47] Liu, Y., Cheng, J., Liu, H., Deng, Y., Wang, J., Xu, F.,Chemom. Intell. Lab. Syst. 2017, 162,142-148.

[48] Schober, D., Jacob, D., Wilson, M., Cruz, J. A., Marcu,A., Grant, J. R., Moing, A., Deborde, C., de Figueiredo, L.F., Haug, K., Rocca-Serra, P., Easton, J., Ebbels, T. M. D.,Hao, J., Ludwig, C., Gunther, U. L., Rosato, A., Klein, M.S., Lewis, I. A., Luchinat, C., Jones, A. R., Grauslys, A.,Larralde, M., Yokochi, M., Kobayashi, N., Porzel. A., Grif-fin, J. L., Viant, M. R., Wishart, D. S., Steinbeck, C., Salek,R. M., Neumann, S., Anal. Chem. 2018, 90,649–656.

[49] Rodriguez-Martinez, A., Posma, J. M., Ayala, R., Neves,A. L., Anwar, M., Petretto, E., Emanueli, C., Gauguier,D., Nicholson, J. K., Dumas, M. E., Bioinformatics 2017,btx477.

[50] Beirnaert, C., Meysman, P., Vu, T. N., Hermans, N.,Apers, S., Pieters, L., Covaci, A., Laukens, K., bioRxiv2017, 138503.

[51] Martin, M., Legat, B., Leenders, J., Vanwinsberghe, J.,Rousseau, R., Boulanger, B., Eilers, P. H., De Tullio, P.,Govaerts, B., No. UCL-Universite Catholique de Lou-vain, 2017.

[52] Margueritte, L., Markov, P., Chiron, L., Starck, J. P., Bour-jot, M. A., Delsuc, M. A., Magn. Reson. Chem. 2017,https://doi.org/10.1002/mrc.4683

[53] Dashti, H., Westler, W. M., Tonelli, M., Wedell, J. R.,Markley, J. L., Eghbalnia, H. R., Anal. Chem. 2017, 89,12201–12208.

[54] Tian, H., Li, B., Shui, G., J. Anal. Test 2017, 1, 187–192.

[55] Burgess, K. E. V., Borutzki, Y., Rankin, N., Daly, R., Jour-dan, F., J. Chromatogr. B 2017, 1071, 68–74

[56] Wen, B., Mei, Z., Zeng, C., Liu, S., BMC Bioinformatics2017, 18, 183.

[57] Wang, Y., Wang, X., Zeng, X., Metabolomics 2017, 13,116.

[58] Edmands, W. M., Petrick, L., Barupal, D. K., Scalbert, A.,Wilson, M. J., Wickliffe, J. K., Rappaport, S. M., Anal.Chem. 2017 89, 3919–3928.

[59] Sakurai, N., Narise, T., Sim, J. S., Lee, C. M., Ikeda, C.,Akimoto, N., Kanaya, S., Bioinformatics 2017, btx649.https://doi.org/10.1093/bioinformatics/btx649

[60] Naake, T., Gaquerel, E., Bioinformatics 2017, 33,2419–2420.

[61] Zhou, Z., Xiong, X., Zhu, Z.J., Bioinformatics 2017, 33,2235–2237.

[62] Zhou, Z., Tu, J., Xiong, X., Shen, X., Zhu, Z.J., Anal.Chem. 2017, 89, 9559–9566.

[63] Ma, J., Casey, C. P., Zheng, X., Ibrahim, Y. M., Wilkins, C.S., Renslow, R. S., Thomas, D. G., Payne, S. H., Monroe,

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

922 B. B. Misra Electrophoresis 2018, 39, 909–923

M. E., Smith, R. D., Teeguarden, J. G., Baker, E. S., Metz,T. O., Bioinformatics 2017, 33, 2715–2722.

[64] Niedenfuhr, S., ten Pierick, A., van Dam, P. T., Suarez-Mendez, C. A., Noh, K., Wahl, S. A., Biotechnol. Bioeng.2016, 113, 1137–1147.

[65] Bueschl, C., Kluger, B., Neumann, N. K., Doppler, M.,Maschietto, V., Thallinger, G. G., Meng-Reiterer, J.,Krska, R., Schuhmacher, R., Anal. Chem. 2017, 89,9518–9526.

[66] Wills, J., Edwards-Hicks, J., Finch, A., Anal. Chem. 2017,89, 9616–9619.

[67] Su, X., Lu, W., Rabinowitz, J. D., Anal. Chem. 2017, 89,5940–5948.

[68] Wei, R., Wang, J., Su, M., Jia, E., Chen, T., Ni, Y., bioRxiv2017, 171967.

[69] Wei, R., Wang, J., Jia, E., Chen, T., Ni, Y., Jia, W., bioRxiv2017, 177410.

[70] Li, L., Ren, W., Kong, H., Zhao, C., Zhao, X., Lin,X., Lu, X., Xu, G., Anal. Chim. Acta. 2017, 990, 96–102.

[71] Boysen, A. K., Heal, K. R., Carlson, L. T., Ingalls, A. E.,Anal. Chem. 2018, 90, 1363–1369.

[72] Thonusin, C., IglayReger, H. B., Soni, T., Rothberg, A.E., Burant, C. F., Evans, C. R., J. Chromatogr. A 2017,1523, 265–274.

[73] Delabriere, A., Hohenester, U. M., Colsch, B., Junot, C.,Fenaille, F., Thevenot, E. A., Bioinformatics 2017, 33,3767–3775.

[74] Systematic Error Removal using Random Forest(SERRF) for large-scale untargeted metabolomicshttp://www.serrf.fiehnlab.ucdavis.edu, 2017.

[75] Brunius, C., Pedersen, A., Malmodin, D., Karlsson, B.G., Andersson, L. I., Tybring, G., Landberg, R., Bioinfor-matics 2017, 33, 3567–3574.

[76] Antonelli, J., Claggett, B., Henglin, M., Watrous, J. D.,Lehmann, K. A., Hushcha, P., Demler, O., Mora, S., Ni-iranen, T., Pereira, A. C., Jain, M., arXiv preprint 2017,arXiv:1710.03436.

[77] Schymanski, E. L., Jeon, J., Gulde, R., Fenner, K., Ruff,M., Singer, H. P., Hollender, J., Environ. Sci. Technol.2014, 48, 2097–2098.

[78] Kind, T., Tsugawa, H., Cajka, T., Ma, Y., Lai, Z., Mehta,S. S., Wohlgemuth, G., Barupal, D. K., Showalter, M.R., Arita, M., Fiehn, O., Mass Spectrom. Rev. 2017,https://doi.org/10.1002/mas.21535.

[79] Lawson, T. N., Weber, R. J., Jones, M. R., Chetwynd,A. J., Rodrıguez-Blanco, G., Di Guida, R., Viant, M. R.,Dunn, W. B., Anal. Chem. 2017, 89, 2432–2439.

[80] Mahieu, N. G., Patti, G. J., Anal. Chem. 2017, 89,10397–10406.

[81] Duhrkop, K., Bocker, S., J. Cheminform. 2016, 8, 5.

[82] Aguilar-Mogas, A., Sales-Pardo, M., Navarro, M.,Guimera, R., Yanes, O., Anal. Chem. 2017 89,3474–3482.

[83] Wandy, J., Zhu, Y., van der Hooft, J. J., Daly, R., Barrett,M. P., Rogers, S., Bioinformatics 2017, btx582.

[84] DIMEdb, https://dimedb.ibers.aber.ac.uk/, 2017

[85] MoNA, http://mona.fiehnlab.ucdavis.edu/, 2017.

[86] Palmer, A., Phapale, P., Fay, D., Alexandrov, T., bioRxiv2017, 170571.

[87] Alden, N., Krishnan, S., Porokhin, V., Raju, R., McE-learney, K., Gilbert, A., Lee, K., Anal. Chem. 2017, 89,13097–13104

[88] Rohart, F., Gautier, B., Singh, A., LeCao, K. A., PLoS Comput. Biol. 2017,https://doi.org/10.1371/journal.pcbi.1005752

[89] Ni, Z., Angelidou, G., Lange, M., Hoffmann, R., Fe-dorova, M., Anal. Chem. 2017, 89, 8800–8807.

[90] CEU Mass Mediator, http://ceumass.eps.uspceu.es/2017.

[91] Hartler, J., Triebl, A., Ziegl, A., Trotzmuller, M., Rech-berger, G. N., Zeleznik, O. A., Zierler, K. A., Torta, F.,Cazenave-Gassiot, A., Wenk, M. R., Fauland, A., Nat.Methods 2017, 14, 1171–1174.

[92] Ni, Z., Angelidou, G., Hoffmann, R., Fedorova, M., Sci.Rep. 2017, 7, 15138.

[93] Guitton, Y., Tremblay-Franco, M., Le Corguille, G., Mar-tin, J. F., Petera, M., Roger-Mele, P., Delabriere, A.,Goulitquer, S., Monsoor, M., Duperier, C., Canlet, C.,Int. J. Biochem. Cell Biol. 2017, 93, 89–101.

[94] MetaboAnalystR, https://github.com/xia-lab/MetaboAnalystR, 2017.

[95] Wanichthanarak, K., Fan, S., Grapov, D., Barupal, D. K.,Fiehn, O., PLoS One 2017, 12, e0171046.

[96] Gloaguen, Y., Morton, F., Daly, R., Gurden, R., Rogers,S., Wandy, J., Wilson, D., Barrett, M., Burgess, K., Bioin-formatics 2017, 33, 4007–4009.

[97] da Veiga Leprevost, F., Gruning, B. A., Alves Aflitos,S., Rost, H. L., Uszkoreit, J., Barsnes, H., Vaudel, M.,Moreno, P., Gatto, L., Weber, J., Bai, M., Bioinformatics2017, 33, 2580–2582.

[98] Slenter, D. N., Kutmon, M., Hanspers, K., Riutta, A.,Windsor, J., Nunes, N., Melius, J., Cirillo, E., Coort, S. L.,Digles, D., Ehrhart, F., Nucleic Acids Res. 2017, gkx1064.

[99] Wishart, D. S., Feunang, Y. D., Marcu, A., Guo, A. C.,Liang, K., Vazquez-Fresno, R., Sajed, T., Johnson, D., Li,C., Karu, N., Sayeeda, Z., Lo, E., Assempour, N., Ber-janskii, M., Singhal, S., Arndt, D., Liang, Y., Badran, H.,Grant, J., Serra-Cayuela, A., Liu, Y., Mandal, R., Neveu,V., Pon, A., Knox, C., Wilson, M., Manach, C., Scalbert,A., Nucleic Acids Res. 2017, gkx1089.

[100] Larralde, M., Lawson, T. N., Weber, R. J., Moreno, P.,Haug, K., Rocca-Serra, P., Viant, M. R., Steinbeck, C.,Salek, R. M., Bioinformatics 2017, 33, 2598–2600.

[101] Lyutvinskiy, Y., Watrous, J.D., Jain, M., Nilsson, R., Anal.Chem. 2017, 89, 5713–5718.

[102] mzR, https://bioconductor.org/packages/release/bioc/html/mzR.html 2017.

[103] Cai, Q., Alvarez, J. A., Kang, J., Yu, T., J. Proteome. Res.2017, 16, 1261–1269.

[104] Hunter, A., Dayalan, S., De Souza, D., Power, B., Lorri-mar, R., Szabo, T., Nguyen, T., O’Callaghan, S., Hack, J.,Pyke, J., Nahid, A., Metabolomics 2017, 13, 14.

[105] Wang, Q., Xu, R., J. Biomed. Inform. 2017, 71, 222–228.

[106] Chazalviel, M., Frainay, C., Poupin, N., Vinson, F., Mer-let, B., Gloaguen, Y., Cottret, L., Jourdan, F., Bioinfor-matics 2017, btx588.

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com

Electrophoresis 2018, 39, 909–923 General 923

[107] Brockmoller, T., Ling, Z., Li, D., Gaquerel, E., Baldwin, I.T., Xu, S., BMC Genomics 2017, 18, 79.

[108] Perez-Riverol, Y., Bai, M., da Veiga Leprevost, F., Squiz-zato, S., Park, Y. M., Haug, K., Carroll, A. J., Spalding, D.,Paschall, J., Wang, M., del-Toro, N., Nature Biotechnol.2017, 35, 406–409.

[109] Ramirez-Gaona, M., Marcu, A., Pon, A., Guo, A. C.,Sajed, T., Wishart, N. A., Karu, N., Djoumbou Feunang,Y., Arndt, D., Wishart, D. S., Nucleic Acids Res. 2017, 45,D440-D445.

[110] Guhlin, J., Silverstein, K.A., Zhou, P., Tiffin, P., Young,N.D., BMC Bioinformatics 2017, 18, 367.

[111] Garg, N., Sethupathy, A., Tuwani, R., NK, R., Dokania,S., Iyer, A., Gupta, A., Agrawal, S., Singh, N., Shukla,S., Kathuria, K., Nucleic Acids Res. 2017, gkx957.

[112] Huang, W., Brewer, L. K., Jones, J. W., Nguyen, A.T., Marcu, A., Wishart, D. S., Oglesby-Sherrouse, A.G., Kane, M. A., Wilks, A., Nucleic Acids Res. 2017,gkx1061.

[113] Zhai, P., Yang, L., Guo, X., Wang, Z., Guo, J., Wang, X.,Zhu, H., BMC bioinformatics 2017, 18, 434.

[114] du Sert, N. P., Bamsey, I., Bate, S. T., Berdoy, M., Clark,R. A., Cuthill, I., Fry, D., Karp, N. A., Macleod, M., Moon,L., Stanford, S. C., PLoS Biol. 2017, 15, e2003779.

[115] Ataman, M., Hatzimanikatis, V., PLoS Comput Biol.2017, 13, e1005513.

[116] Rosenthal, S. B., Len, J., Webster, M., Gary, A., Birm-ingham, A., Fisch, K.M., Bioinformatics 2017, btx581.

[117] ElMaven, https://elucidatainc.github.io/ElMaven/, 2017.

[118] Erbilgin, O., Ruebel, O., Louie, K. B., Trinh, M., deRaad, M., Wildish, T., Udwary, D. W., Hoover, C. A.,Deutsch, S., Northen, T. R., Bowen, B. P., bioRxiv 2017,204362.

[119] Zacharias, H. U., Rehberg, T., Mehrl, S., Richt-mann, D., Wettig, T., Oefner, P. J., Spang, R., Gron-wald, W., Altenbuchinger, M., arXiv preprint 2017,arXiv:1703.07724.

[120] Del Carratore, F., Jankevics, A., Eisinga, R., Heskes,T., Hong, F., Breitling, R., Bioinformatics 2017, 33,2774–2775.

[121] Yu, N., Lobue, P. A., Cao, X., Limbach, P. A., Anal. Chem.2017, 89, 10744–10752.

[122] Liu, G., Cheng, K., Lo, C. Y., Li, J., Qu, J., Neelamegham,S., Mol. Cell Proteomics 2017, 16, 2032–2047.

[123] Inacio, B., Ferreira, J. D., Couto, F. M., International Con-ference on Practical Applications of Computational Bi-ology & Bioinformatics, Springer, Cham., pp. 197–204.

[124] Lekschas, F., Gehlenborg, N., Bioinformatics 2017,btx739.

[125] Gruning, B., Dale, R., Sjodin, A., Rowe, J., Chapman, B.A., Tomkins-Tinch, C. H., Valieris, R., Koster, J., bioRxiv2017, 207092.

[126] Ranzinger, R., Weatherly, B., Arpinar, S., Khan, S.,Porterfield, M., Tiemeyer, M., York, W. S., GRITSToolbox-A software system for the archival, process-ing and interpretation of glycomics data, 2015, OxfordUniv Press Inc., NC, USA, pp. 1275-1275.

[127] Christensen, C. D., Hofmeyr, J. H. S., Rohwer, J. M.,Bioinformatics 2017, btx567.

[128] Barupal, D. K., Fiehn, O., Sci. Rep. 2017, 7, 14567.

[129] NormalizeMets, https://cran.r-project.org/web/packa-ges/NormalizeMets/index.html, 2017

[130] Sandusky, P. O., J. Chem. Educ. 2017, 94, 1324–1328.

[131] Dunn, W. B., Broadhurst, D. I., Edison, A., Guillou, C.,Viant, M. R., Bearden, D. W., Beger, R. D., Metabolomics2017, 13, 50.

C© 2018 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim www.electrophoresis-journal.com