ONTOLOGY SOURCE REFERENCE
Term Source Name	OBI	MTBLS	MSIO	EFO	NCIT	roleAFRL	DOID	CHMO	
Term Source File	http://data.bioontology.org/ontologies/OBI	https://www.ebi.ac.uk/metabolights/							http://purl.obolibrary.org/obo/ncit.owl
Term Source Version	29	1.0							18.10e
Term Source Description	Ontology for Biomedical Investigations	Metabolights Ontology							NCI Thesaurus OBO Edition
INVESTIGATION
Investigation Identifier	MTBLS719
Investigation Title	Investigation
Investigation Description	
Investigation Submission Date	
Investigation Public Release Date	2022-03-01
Comment[Created With Configuration]	/Volumes/GoogleDrive/My Drive/ISAcreatorMetaboLights/Configurations/MetaboLightsConfig20150707
Comment[Last Opened With Configuration]	MetaboLightsConfig20150707
INVESTIGATION PUBLICATIONS
Investigation PubMed ID
Investigation Publication DOI
Investigation Publication Author List
Investigation Publication Title
Investigation Publication Status
Investigation Publication Status Term Accession Number
Investigation Publication Status Term Source REF
INVESTIGATION CONTACTS
Investigation Person Last Name
Investigation Person First Name
Investigation Person Mid Initials
Investigation Person Email
Investigation Person Phone
Investigation Person Fax
Investigation Person Address
Investigation Person Affiliation
Investigation Person Roles
Investigation Person Roles Term Accession Number
Investigation Person Roles Term Source REF
STUDY
Study Identifier	MTBLS719
Study Title	Metabolic phenotyping of urine from a dementia research cohort
Study Description	To better understand the molecular mechanisms underpinning physiological variation in human populations, metabolic phenotyping approaches are increasingly being applied to studies involving hundreds and thousands of biofluid samples. Hyphenated ultra-performance liquid chromatography-mass spectrometry (UPLC-MS) has become a fundamental tool for this purpose. However, the seemingly inevitable need to analyze large studies in multiple analytical batches for UPLC-MS analysis poses a challenge to data quality which has been recognized in the field. Herein, we describe in detail a fit-for-purpose UPLC-MS platform, method set, and sample analysis workflow, capable of sustained analysis on an industrial scale and allowing batch-free operation for large studies. Using complementary reversed-phase chromatography (RPC) and hydrophilic interaction liquid chromatography (HILIC) together with high resolution orthogonal acceleration time-of-flight mass spectrometry (oaTOF-MS), exceptional measurement precision is exemplified with independent epidemiological sample sets of approximately 650 and 1000 participant samples. Evaluation of molecular reference targets in repeated injections of pooled quality control (QC) samples distributed throughout each experiment demonstrates a mean retention time relative standard deviation (RSD) of <0.3% across all assays in both studies and a mean peak area RSD of <15% in the raw data. To more globally assess the quality of the profiling data, untargeted feature extraction was performed followed by data filtration according to feature intensity response to QC sample dilution. Analysis of the remaining features within the repeated QC sample measurements demonstrated median peak area RSD values of <20% for the RPC assays and <25% for the HILIC assays. These values represent the quality of the raw data, as no normalization or feature-specific intensity correction was applied. While the data in each experiment was acquired in a single continuous batch, instances of minor time-dependent intensity drift were observed, highlighting the utility of data correction techniques despite reducing the dependency on them for generating high quality data. These results demonstrate that the platform and methodology presented herein is fit-for-use in large scale metabolic phenotyping studies, challenging the assertion that such screening is inherently limited by batch effects. Details of the pipeline used to generate high quality raw data and mitigate the need for batch correction are provided.
Study Submission Date	2018-07-25
Study Public Release Date	2022-03-01
Study File Name	s_NPC.txt
STUDY DESIGN DESCRIPTORS
Study Design Type	untargeted metabolites	ultra-performance liquid chromatography-mass spectrometry	nuclear magnetic resonance spectroscopy	Urine global profiling	Alzheimer's Disease
Study Design Type Term Accession Number	http://www.ebi.ac.uk/metabolights/ontology/MTBLS_000279	http://purl.obolibrary.org/obo/CHMO_0000715	http://purl.obolibrary.org/obo/CHMO_0000591	http://www.ebi.ac.uk/metabolights/ontology/MTBLS_000281	http://purl.obolibrary.org/obo/NCIT_C2866
Study Design Type Term Source REF	MTBLS	CHMO	CHMO	MTBLS	NCIT
STUDY PUBLICATIONS
Study PubMed ID	27479709
Study Publication DOI	10.1021/acs.analchem.6b01481
Study Publication Author List	Lewis Matthew R., Pearce Jake T. M., Spagou Konstantina, Green Martin, Dona Anthony C., Yuen Ada H. Y., David Mark, Berry David J., Chappell Katie, Horneffer-van der Sluis Verena, Shaw Rachel, Lovestone Simon, Elliott Paul, Shockcor John, Lindon John C., Cloarec Olivier, Takats Zoltan, Holmes Elaine, Nicholson Jeremy K.
Study Publication Title	Development and Application of Ultra-Performance Liquid Chromatography-TOF MS for Precision Large Scale Urinary Metabolic Phenotyping
Study Publication Status	Published
Study Publication Status Term Accession Number	http://www.ebi.ac.uk/efo/EFO_0001796
Study Publication Status Term Source REF	EFO
STUDY FACTORS
Study Factor Name	Gender	Age	Sample Dilution
Study Factor Type	Gender	Age	Sample Dilution
Study Factor Type Term Accession Number	http://purl.obolibrary.org/obo/NCIT_C17357	http://purl.obolibrary.org/obo/NCIT_C25150	http://purl.obolibrary.org/obo/NCIT_C178974
Study Factor Type Term Source REF	NCIT	NCIT	NCIT
STUDY ASSAYS
Study Assay File Name	a_MTBLS719_LC-MS_positive_reverse-phase_metabolite_profiling.txt	a_MTBLS719_NMR___metabolite_profiling.txt	a_MTBLS719_LC-MS_negative_reverse-phase_metabolite_profiling.txt	a_MTBLS719_LC-MS_positive_hilic_metabolite_profiling.txt
Study Assay Measurement Type	metabolite profiling	metabolite profiling	metabolite profiling	metabolite profiling
Study Assay Measurement Type Term Accession Number	http://purl.obolibrary.org/obo/OBI_0000366	http://purl.obolibrary.org/obo/OBI_0000366	http://purl.obolibrary.org/obo/OBI_0000366	http://purl.obolibrary.org/obo/OBI_0000366
Study Assay Measurement Type Term Source REF	OBI	OBI	OBI	OBI
Study Assay Technology Type	mass spectrometry	NMR spectroscopy	mass spectrometry	mass spectrometry
Study Assay Technology Type Term Accession Number	http://purl.obolibrary.org/obo/OBI_0000470	http://purl.obolibrary.org/obo/OBI_0000623	http://purl.obolibrary.org/obo/OBI_0000470	http://purl.obolibrary.org/obo/OBI_0000470
Study Assay Technology Type Term Source REF	OBI	OBI	OBI	OBI
Study Assay Technology Platform	Liquid Chromatography MS - positive - reverse phase	Nuclear Magnetic Resonance (NMR) -	Liquid Chromatography MS - negative - reverse phase	Liquid Chromatography MS - positive - hilic
STUDY PROTOCOLS
Study Protocol Name	Sample collection	Extraction	Chromatography	Mass spectrometry	Data transformation	Metabolite identification	NMR sample	NMR spectroscopy	NMR assay
Study Protocol Type	Sample collection	Extraction	Chromatography	Mass spectrometry	Data transformation	Metabolite identification	NMR sample	NMR spectroscopy	NMR assay
Study Protocol Type Term Accession Number									
Study Protocol Type Term Source REF									
Study Protocol Description	<p>The Alzheimer’s disease study consists of a series of serum, plasma and urine biofluid samples collected as part of the AddNeuroMed and ART/DCR study consortia<strong>[1]</strong>, which follow patients longitudinally with the aim of identifying biomarkers of neurocognitive decline and Alzheimer’s disease. The samples analysed by LC-MS and NMR, and for which data is provided in this MetaboLights study, are baseline spot urine samples (first sample collected after&nbsp;recruitment to the study). Patients were asked to fast for 2 h prior to urine sample collection, and no samples in this set were freeze-thawed before use in this trial.</p><p><br></p><p>The cohort consists of 2 broad groups of subjects. Subjects with <strong>AD </strong>were collected from the South London and Maudsley NHS Foundation Trust. Subjects with MCI and unaffected <strong>controls </strong>were ascertained through primary care. In brief, participating GP clinics contacted all registered patients over the age of 70 and patients were invited to attend the practice for an assessment of memory. Screening assessment consisted of information and consent processes and a screening assessment for dementia. Subjects with normal cognition were entered into the study as unaffected controls, patients with impaired cognition or subjective memory impairment then underwent a full assessment consisting of cognition, behaviour and function together with an extensive clinical interview. Subjects meeting Petersen criteria for amnestic Mild Cognitive Impairment and NINCDS-ADRDA criteria for possible or probable Alzheimer’s disease were classified accordingly. A proportion of subjects have had MRI, and in some cases, serial MRI. All subjects were assessed annually – <strong>controls </strong>by a standardised telephone interview for cognition (TICS-M) and MCI and <strong>AD </strong>groups using a full interview including cognitive assessment. <strong>Control </strong>subjects receive full interview assessment if cognitive score suggests deterioration and possible conversion to MCI. Conversion from MCI to dementia is assessed by cognitive and other deterioration and by review by an experienced clinician. Detailed information about this cohort and clinical metadata can be found in the previous published paper<strong>[1] </strong>and the ANMERGE repository<strong>[2]</strong>, available via the Sage BioNetworks portal (<a href='https://doi.org/10.7303/syn22252881' rel='noopener noreferrer' target='_blank'>https://doi.org/10.7303/syn22252881</a>).</p><p><br></p><p>The Alzheimer’s disease study samples came from the AddNeuroMed and ART/DCR studies with ethical approval from boards in the respective study site institutions across Europe.</p><p><br></p><p><strong>Sample handling and aliquotting</strong></p><p>Urine samples were received on dry ice and immediately stored at -80 °C. Meta-data variables essential to the design of the study were identified and used to determine sample analysis orders that were uncorrelated with those factors using in-house software, avoiding the potential for confounding critical clinical variables with analytical run order effects. For each study, this order was preloaded into an in-house laboratory information management system (LIMS). Samples were then sorted into sets of 80 using aluminium racks (9x9 position format) cooled by direct contact with dry ice to ensure that all samples remained frozen for the duration of the sorting exercise. Completed sets were stored at -80 °C until required for subsequent steps.</p><p><br></p><p>Once sorted, individual urine sample sets were reformatted into 96-well plates. This step required the overnight thawing of a set of 80 samples at 4 °C, mixing of each individual sample by vortexing, and transference of the liquid contents of each sample to a well within a 96-deep-well polypropylene plate (2 mL, Eppendorf). During this process, each sample was checked against a LIMS-generated map of sample identifiers and positions in the 96-well plate to validate the batch composition and sample order. Study samples were added only to the wells within the first 10 columns of rows A through H; columns 11 and 12 of the plate were reserved for the later addition of pooled quality control (<strong>QC</strong>) reference samples. The plate was sealed with a silicone cap mat and centrifuged at 3,486 × g for 10 min at 4 °C using an Eppendorf 5810R centrifuge equipped with an A-2-DWP-AT rotor for the removal of suspended particulate matter such as cryoprecipitate, cellular debris, or crystalline material. The supernatant was aspirated and dispensed in pre-programmed volumes among 96-well polypropylene plates using an 8-channel 15-1200 μL Eppendorf Xplorer Plus pipette with an initial discarding step into an 800 mL beaker to be used as <strong>NMR </strong>Study Reference. The volume dispensed in plates included 600 μL to a 2.0 mL plate (<strong>NMR</strong>), 150 μL to a 0.5 mL plate (reversed-phase chromatography, <strong>RPC</strong>), 50 μL to a 0.5 mL plate (hydrophilic interaction liquid chromatography, <strong>HILIC</strong>) and 50 μL to a 0.3 mL plate (study reference, <strong>SR</strong>, sample). All plates were sealed with a silicone cap mat and returned to -80 °C.</p><p><br></p><p><strong>Preparation of Method Reference (MR) and Internal Standards (IS) mixtures</strong></p><p>Mixtures of reference chemicals were generated for both <strong>RPC </strong>and <strong>HILIC </strong>methods, enabling the targeted assessment of instrument stability (e.g.chromatographic retention, mass accuracy, and signal intensity) and data quality. <strong>IS </strong>are added to all samples in the study and<strong> MR</strong> standards are also added to all <strong>QC </strong>samples. L-phenylalanine13C9, 15N (from Sigma-Aldrich (Steinheim, Germany)) and N-benzoyl-d5-glycine (hippuric acid, from QMX (Essex, U.K.)) were used in both chromatographic assay types as a 2-point IS. The RPC <strong>MR </strong>was composed of 8 reference standards from Sigma-Aldrich and QMX: L-glutamic acid-13C5; L-isoleucine-13C6,15N; L-leucine-13C6, L-tryptophan-13C11,15N2; L-glutamine-13C5; creatinine-(methyl-d3), cytidine-5,6-d2; and benzoic acid-phenyl-13C6. 2 of these standards were determined to not be suitable for analysis of <strong>RPC+</strong> data quality (L-glutamine13C5 and benzoic acid-phenyl13C6), while 1 was not suitable for analysis <strong>RPC-</strong> data (creatinine-(methyl-d3), cytidine-5,6-d2). The <strong>HILIC MR </strong>was composed of 6 reference standards from Sigma-Aldrich, QMX and CDN isotopes (Quebec, Canada): adenosine-2-d1; adenine-2-d1; 2-aminoethane-d4-sulfonic acid(taurine-d4); creatine-(guanidino-13C)monohydrate; L-arginine-13C6; and uracil-2-13C,15N2.</p><p><br></p><p><strong>Preparation of Quality Control (QC) samples</strong></p><p>QC samples were analysed at regular intervals throughout the run and utilized to support the assessment of data quality within each experiment for LC-MS and NMR analysis.</p><p><br></p><p><strong>Long Term Reference (LTR)</strong></p><p>A pooled <strong>Long Term Reference</strong> (<strong>LTR</strong>) urine sample is maintained by the laboratory and is used as an independent sample reference throughout all molecular profiling studies. To create this material, 78 individual urine voids were collected in a single day from volunteer subjects. No screening criteria were used to assess the health status of the donors. All samples were stored at 4 °C from collection until the following day when all samples were combined and aliquoted by the following procedure. Individual samples were centrifuged at 4 °C for 15 min at 3,214 × g and their supernatants combined in a 20 L polypropylene vessel. The pooled sample was homogenized in the vessel by gentle stirring using a Teflon-coated magnetic stir bar and magnetic stir plate. Aliquots were created by multiple iterations of dispensing 1 L of pooled urine supernatant to a 1 L glass bottle and aliquoting that material into 15 mL polypropylene conical centrifuge tubes (Corning) using an Eppendorf 5-25 mL Varispenser plus. This process was repeated until all of the reference material was aliquoted. All aliquots were stored at -80 °C until required for further preparation. A sufficient volume of LTR sample (approximately 2 mL per 80 study samples) was thawed for <strong>RPC </strong>and <strong>HILIC </strong>assays and supplemented with the relevant <strong>RPC </strong>and <strong>HILIC MR </strong>mixtures in a 2:1 LTR:MR ratio, homogenized, and aliquoted to 1.7 mL polypropylene microcentrifuge tubes. All LTR materials were frozen at -80 °C until required for analytical plate preparation.</p><p><br></p><p><strong>Study Reference (SR)</strong></p><p>A pooled <strong>Study Reference </strong>(<strong>SR</strong>)<strong> </strong>sample representing the mean of samples within the Alzheimer’s study was prepared. For <strong>NMR </strong>analysis the <strong>SR </strong>was created by pooling a small volume of each sample in the study into an 800 mL beaker during the sampling reformatting step. The beaker was frozen overnight and homogenised at the end of the reformatting process, then subsequently split into 5 mL aliquots and taken back to -80 °C. 2 aliquots of 600 μL were added to each <strong>NMR </strong>plate before analysis. For <strong>LC-MS</strong> analysis, the <strong>SR </strong>was created by thawing the contents of all <strong>SR </strong>sample plates (with 50 μL of each sample) at 4°C and combining their contents in an 800 mL glass beaker. Optionally, individual samples determined to be unsuitable for <strong>UPLC-MS</strong> analysis (by sample pre-screening via <strong>NMR </strong>spectroscopy) were removed from the individual <strong>SR </strong>sample plate wells prior to pooling in order to avoid incorporation of severely contaminated or outlying samples which may compromise the <strong>SR</strong>. The pooled sample was homogenized by swirling and portioned in an approximate 3:1 ratio for use with <strong>RPC </strong>and <strong>HILIC </strong>assays respectively. The <strong>RPC </strong>and <strong>HILIC </strong>sub-fractions were supplemented with the relevant <strong>RPC </strong>and <strong>HILIC MR</strong> mixtures in a 2:1 <strong>SR:MR</strong> ratio, homogenized, and aliquoted to 1.7 mL polypropylene microcentrifuge tubes. All <strong>SR </strong>materials were frozen at -80 °C until required for analytical plate preparation.</p><p><br></p><p><strong>Dilution series</strong></p><p>A dilution series of <strong>SR </strong>samples was prepared for each assay type from thawed <strong>SR </strong>in the same manner described above except that they were prepared in bulk volume in Eppendorf microcentrifuge tubes and the products were stored in polypropylene sample vials for analysis supporting multiple injections from the same stock. Prior to and immediately following the analysis of each experiment’s samples, dilution series samples were analysed by repeated measurement aimed at weighting the ends of the series (10,10,5,3,3,5, and 10 consecutive replicate injections of the 1%, 10%, 20%, 40%, 60%, 80%, and 100% <strong>SR </strong>samples respectively).</p><p><br></p><p><strong>Refs:</strong></p><p><strong>[1] </strong>Lovestone S, Francis P, Kloszewska I, Mecocci P, Simmons A, Soininen H, Spenger C, Tsolaki M, Vellas B, Wahlund LO, Ward M; AddNeuroMed Consortium. AddNeuroMed--the European collaboration for the discovery of novel biomarkers for Alzheimer's disease. Ann N Y Acad Sci. 2009 Oct;1180:36-46. doi:10.1111/j.1749-6632.2009.05064.x. PMID:19906259</p><p><strong>[2]</strong> Birkenbihl C, Westwood S, Shi L, Nevado-Holgado A, Westman E, Lovestone S; AddNeuroMed Consortium, Hofmann-Apitius M. ANMerge: A Comprehensive and Accessible Alzheimer's Disease Patient-Level Dataset. J Alzheimers Dis. 2021;79(1):423-431. doi: 10.3233/JAD-200948. PMID:33285634; PMCID:PMC7902946</p>	<p><strong>LC-MS sample preparation</strong></p><p>Individual aliquot plates were removed from -80 °C storage approximately 3 h prior to their scheduled analysis and thawed at 4 °C. All study samples were diluted with ultrapure water (75 μL for RPC preparation and 25 μL for HILIC preparation), commensurate with the volume of QC sample dilution caused by the addition of MR. LTR stock and SR stock aliquots were thawed and their contents distributed among the wells of columns 11 and 12, respectively (225 μL for RPC analysis and 75 μL for HILIC analysis). Finally, the IS solution was added to all samples (75 μL for RPC preparation and 25 μL for HILIC preparation). Sample plates for RPC analysis were sealed with a silicone cap mat, mixed for 1 min at 4 °C using a MixMate (Eppendorf) operating at 850 rpm, and centrifuged for 10 min at 3,486 × g. The supernatant was split between 2 analytical microwell plates, 1 for positive and 1 for negative ion mode UPLC-MS analysis. Acetonitrile (ACN) was added to samples for HILIC analysis in a 3:1 ACN:sample ratio to better match the initial solvent conditions of the chromatographic method. The sample plate was then sealed with Thermo-Seal heat sealing foil sheets using an ALPS 50 V-Manual Heat Sealer (Thermo SCIENTIFIC) prior to mixing for 1 minute at 4 °C using a MixMate (Eppendorf) operating at 850 rpm. The plates were subsequently centrifuged for 10 min at 3486 × g to remove any residual protein precipitated by this addition of organic solvent, and a 150 μL volume of the supernatant was decanted to a microplate for analysis. Once prepared, plates were loaded into the UPLC sample manager where they were held at 4 °C until analysis.</p>	<p>All assays were conducted on independent but identical UPLC-MS systems. The LC component was an ACQUITY UPLC (Waters Corp., Milford, MA, USA) composed of a binary solvent manager and column heater/cooler module. The sample handling component was a Waters 2777C sample manager (Waters Corp., Milford, MA, USA) equipped with a 25 μL Hamilton syringe, a 2 μL loop used for full-loop injections (with a 5-fold overfill were used, aspirating and injecting 10 µL of prepared sample), and a 3-drawer sample chamber thermo-stated at 4 °C with a constant flow of dry nitrogen gas. Seal wash consisted of isopropanol:water (1:9, v/v), weak needle wash isopropanol:water (1:9, v/v) for RPC and ACN:water (3:1, v/v)&nbsp;for HILIC and the strong needle wash was 100% isopropanol. All reagents were of LCMS grade from Sigma-Aldrich (Steinheim, Germany) (formic acid, ACN and isopropanol) and Fisher Scientific (Hampton, NH, USA) (water).</p><p><br></p><p><strong>Reverse Phase</strong></p><p>For RPC, separation was performed in a 2.1×150 mm HSS T3 column (Waters Corp., Milford, MA, USA) thermostatted at 45 °C was used with a mobile phase flow rate of 0.6 mL/min. The solvent system chosen was water (A) and acetonitrile (B) each supplemented with 0.1% formic acid. After a 0.1 min isocratic separation at initial conditions (99% A), a linear gradient elution (99% to 45% A in 9.9 min) was applied, followed by a more rapid gradient (45% to 0% A in 0.7 min) to final conditions. In the latter stage, the mobile phase flow rate was simultaneously increased to 1.0 mL/min, allowing faster column washing. Further details on the gradient conditions can be found in the paper associated with this study.</p><p><br></p><p><strong>HILIC</strong></p><p>For HILIC, separation was conducted using a 2.1×150 mm Acquity BEH HILIC column (Waters Corp., Milford, MA, USA) thermostatted at 40 °C. The solvent system was acetonitrile with 0.1% formic acid (A) and 20 mM ammonium formate in water with 0.1%formic acid (B) with a flow rate of 0.6 mL/min. After a 0.1 min isocratic separation at initial conditions (95% A), a 2-stage gradient was conducted. First, a shallow linear gradient between 95% and 80% A was used followed by a more rapid gradient from 80% to 50% A. Following a return to initial composition, the flow rate was increased to 1.0 mL/min to expedite extended equilibration of the chromatographic system. Further details on the gradient conditions can be found in the paper associated with this study.</p><p><br></p>	<p>All assays were conducted on independent but identical UPLC-MS systems. The MS component was a Xevo G2-S oaTOF MS (Waters Corp., Manchester, UK) coupled to the UPLC via a Zspray electrospray ionization (ESI) source. Instrument ion source and ion guide settings were as follows for all assays, unless specified otherwise: capillary = 1.5/1.0 kV for RPC ESI+, HILIC/RPC ESI-, source temperature 120 °C, desolvation temperature 600 °C, desolvation gas flow 1000 L/h, cone gas flow = 150 L/hr, cone = 20 V, source offset = 80 V, and StepWave 2 offset = 10 V. Data were collected in centroid mode with a scan range of 50-1200 <em>m/z</em> for small molecule analysis, and a scan time of 0.07/0.1s for HILIC/RPC. The angular positioning of the sample probe in relation to the inlet cone was set to 7 mm on the adjustment micrometer.</p><p><br></p><p>For mass accuracy, instrument calibration was performed using sodium formate before each ESI mode acquisition. LockSpray mass correction was performed using a 200 pg/μL leucine enkephalin solution (<em>m/z</em> 556.2771 in ESI+, <em>m/z</em> 554.2615 in ESI-) in 50:50 H2O:ACN solution at a flow rate of 15 μL/min. Lockmass scans were collected every 60 s and averaged over 3 scans.</p><p><br></p><p>All reagents were of LCMS grade from Sigma-Aldrich (Steinheim, Germany) (ACN, formic acid, ammonium formate), Fisher Scientific (Hampton, NH, USA) (water) and Waters Corp (leucine enkephalin).</p>	<p><strong>Mass Spectral Data</strong></p><p>The raw data files were converted from the Waters <strong>.RAW</strong> format to <strong>.mzML</strong> format using the <strong>msconvert tool</strong> from the <strong>ProteoWizard </strong>tookit [1]. <strong>msconvert </strong>was configured to discard scans below an intensity threshold of 100, and only retain scans from the first acquisition function (--filter 'scanEvent 1' --filter 'threshold absolute 100 most-intense').</p><p><br></p><p><strong>Ref</strong>:</p><p><strong>[1] </strong>Chambers MC, Maclean B, Burke R, Amodei D, Ruderman DL, Neumann S, Gatto L, Fischer B, Pratt B, Egertson J, Hoff K, Kessner D, Tasman N, Shulman N, Frewen B, Baker TA, Brusniak MY, Paulse C et al., A cross-platform toolkit for mass spectrometry and proteomics. Nat Biotechnol. 2012 Oct;30(10):918-20. doi:10.1038/nbt.2377. PMID:23051804</p>	<p>Metabolite annotation and identification was performed using a combination of computational tools and experiments. Experiments were performed preferentially on the <strong>Long Term Reference</strong> and <strong>Study Reference</strong> samples, or if the former was not possible, on study samples containing the chemical signal in high abundance.</p><p><br></p><p>We used the Metabolomics Standards Initiative (<strong>MSI</strong>) criteria<strong>[1] </strong>to define the level of confidence in each metabolite’s annotation.&nbsp;Metabolites with an MSI level of 1 were annotated by matching signals to entries in an <strong>in-house database</strong>. The retention time tolerance for feature matching was 5 s for Reversed phase and 15 s for HILIC chromatography, and the <em>m/z</em> tolerance was 15 part-per-million (ppm). </p><p>Additionally, MS/MS spectra and isotopic pattern were manually verified to further confirm identity. We have assigned an MSI level of 2 to annotations were the metabolites were not present in our standards database but for which MS/MS spectra were available in the public spectral databases such as <strong>HMDB[2]</strong>, <strong>Metlin[3]</strong>, <strong>Mass Bank of North America</strong> (<a href='https://massbank.us' rel='noopener noreferrer' target='_blank'>https://massbank.us</a>) and <strong>NIST17</strong>. Finally, metabolites annotated with an MSI level equal to 3 were assigned by matching their MS/MS spectra to spectral information available in published literature and/or predicted using <strong>CFM-ID[4]</strong>, <strong>MAGMa software[5]</strong>, or <strong>MetFrag[6]</strong>.&nbsp;</p><p><br></p><p>For NMR spectroscopy, putative metabolite annotation was performed using<strong> in-house </strong>and <strong>public databases</strong> (<strong>HMDB[2]</strong>, <strong>BioMagResBank[7]</strong> of chemical shifts. When adequate, statistical spectroscopy (<strong>STOCSY</strong>)<strong>[8]</strong> was used to exploit structural correlations between signals and help establish correspondence between different resonances from the same compound, expanded the information available for spectral matching. When possible, assignments were confirmed using the coupling constant J measured experimentally with the J-resolved pulse sequence. For the annotation of some metabolites, a range of 2D NMR spectroscopy experiments including <strong>J-resolved,</strong> <strong>1H,1H-COSY,</strong> <strong>1H</strong>,<strong>1H-TOCSY</strong>, <strong>2D 1H</strong>,<strong>1H-NOESY</strong>, <strong>1H</strong>,<strong>13C-HSQC</strong>, and <strong>1H</strong>,<strong>13C-HMBC</strong> were performed. The relaxation delays were set to 2 s to reduce the experimental time, and all pulse sequences included a pre-saturation period to suppress water signal. The 1H spectral window was set to 12.0 ppm and either 190 or 230 ppm for the 13C windows of <strong>HSQC </strong>or <strong>HMBC </strong>respectively. The centre of the 1H window was centred on the water signal and either at 85 or 105 ppm for the indirect dimension of <strong>HSQC </strong>and <strong>HMBC</strong>, respectively.</p><p>&nbsp;</p><p><strong>Refs</strong>:</p><p><strong>[1]</strong> Sumner LW, Amberg A, Barrett D, Beale MH, Beger R, Daykin CA, Fan TW, Fiehn O, Goodacre R, Griffin JL, Hankemeier T, Hardy N, Harnly J, Higashi R, Kopka J, Lane AN, Lindon JC, Marriott P, Nicholls AW, Reily MD, Thaden JJ, Viant MR. Proposed minimum reporting standards for chemical analysis Chemical Analysis Working Group (CAWG) Metabolomics Standards Initiative (MSI). Metabolomics. 2007 Sep;3(3):211-221. doi:10.1007/s11306-007-0082-2. PMID:24039616</p><p><strong>[2]</strong> Wishart DS, Tzur D, Knox C, Eisner R, Guo AC, Young N, Cheng D, Jewell K, Arndt D, Sawhney S, Fung C, Nikolai L, Lewis M, Coutouly MA, Forsythe I, Tang P, Shrivastava S, Jeroncic K, Stothard P et al. HMDB: the Human Metabolome Database. Nucleic Acids Res. 2007 Jan;35(Database issue):D521-6. doi: 10.1093/nar/gkl923. PMID: 17202168</p><p><strong>[3] </strong>Smith CA, O'Maille G, Want EJ, Qin C, Trauger SA, Brandon TR, Custodio DE, Abagyan R, Siuzdak G. METLIN: a metabolite mass spectral database. Ther Drug Monit. 2005 Dec;27(6):747-51. doi:10.1097/01.ftd.0000179845.53213.39. PMID:16404815</p><p><strong>[4]</strong> Allen F, Pon A, Wilson M, Greiner R, Wishart D. CFM-ID: a web server for annotation, spectrum prediction and metabolite identification from tandem mass spectra. Nucleic Acids Res. 2014 Jul;42(Web Server issue):W94-9. doi:10.1093/nar/gku436. Epub 2014 Jun 3. PMID:24895432</p><p><strong>[5] </strong>Verdegem, D., Lambrechts, D., Carmeliet, P., &amp; Ghesquière, B. (2016). Improved metabolite identification with MIDAS and MAGMa through MS/MS spectral dataset-driven parameter optimization. Metabolomics. https://doi.org/10.1007/s11306-016-1036-3</p><p><strong>[6]</strong> Wolf S, Schmidt S, Müller-Hannemann M, Neumann S. In silico fragmentation for computer assisted identification of metabolite mass spectra. BMC Bioinformatics. 2010 Mar 22;11:148. doi:10.1186/1471-2105-11-148. PMID:20307295</p><p><strong>[7] </strong>Ulrich EL, Akutsu H, Doreleijers JF, Harano Y, Ioannidis YE, Lin J, Livny M, Mading S, Maziuk D, Miller Z, Nakatani E, Schulte CF, Tolmie DE, Kent Wenger R, Yao H, Markley JL. BioMagResBank. Nucleic Acids Res. 2008 Jan;36(Database issue):D402-8. doi: 10.1093/nar/gkm957. Epub 2007 Nov 4. PMID: 17984079</p><p><strong>[8] </strong>Cloarec O, Dumas ME, Craig A, Barton RH, Trygg J, Hudson J, Blancher C, Gauguier D, Lindon JC, Holmes E, Nicholson J. Statistical total correlation spectroscopy: an exploratory approach for latent biomarker identification from metabolic 1H NMR data sets. Anal Chem. 2005 Mar 1;77(5):1282-9. doi:10.1021/ac048630x. PMID:15732908</p>	<p>2 Long Term Reference and 2 Sample Reference samples (600 mL per well) were added to the each 96 well plate in positions C11, G11 and C12, G12 respectively. QC samples were thawed and thoroughly mixed before adding the samples to the relevant well. The rest of the plate was kept frozen by keeping it over dry ice on a metallic tray.&nbsp;</p><p><br></p><p>To prepare samples for 1H NMR spectroscopy, 540 μL of urine and 60 μL of D2O buffer containing 1.5 M of KH2PO4, 5.8 mM of TSP (3-(trimethylsilyl)-2,2,3,3-tetradeuteropropionic acid or TMSP-d4) and 2 mM of NaN3 at pH 7.4 were introduced into 4’’ 5 mm NMR tubes. A 215 Gilson liquid handling robot was used for the preparation. Samples were manually mixed by inverting several times the sealed tubes in the SampleJet racks and immediately loaded onto a refrigerated SampleJet robot (Bruker Coorporation, Germany) and kept at 6 °C until measurement.&nbsp;</p>	<p>Experiments were performed on a Bruker Avance III HD 600 MHz spectrometer working at 14.1 T and equipped with a BBI probe with z-gradients and high order shims and automatic tuning and matching. The probe temperature was set to 300 K. Temperature calibration using a standard methanol sample, assessment of spectrometer performance and water suppression, and calibration of the eretic signal for quantitation was performed before the acquisition of each set of samples as described in the previously published paper<strong>[1]</strong>. <strong>TopSpin v. 3.6</strong> (Bruker Coorporation, Germany) was used for acquisition and automatic processing of all spectra.&nbsp;</p><p><br></p><p><strong>Ref:</strong></p><p><strong>[1] </strong>Dona AC, Jiménez B, Schäfer H, Humpfer E, Spraul M, Lewis MR, Pearce JT, Holmes E, Lindon JC, Nicholson JK. Precision high-throughput proton NMR spectroscopy of human urine, serum, and plasma for large-scale metabolic phenotyping. Anal Chem. 2014 Oct 7;86(19):9887-94. doi:10.1021/ac5025039. Epub 2014 Sep 16. PMID:25180432</p>	<p><strong>Noesy-presat sequence</strong></p><p>The 1H NMR spectra were measured using a specified water suppression pulse sequence (<strong>noesygppr1d</strong> in Bruker database) which employs the first increment of a NOESY pulse sequence with continuous wave irradiation at the water resonance frequency using 25 Hz RF strength during the relaxation delay and also during the mixing time. The sequence has the form –RD –gz,1 –90° –t –90° –tm–gz,2–90° –ACQ, where RD is the relaxation delay (4 s), t is a short delay typically of about 3 μs, 90° represents a 90° RF pulse, tm is the mixing time (10 ms), gz,1 and gz,2 are magnetic field z-gradients both applied for 1 ms, and ACQ is the data acquisition period (2.7 s). Application of the gradients ensures that dispersive residual water signals are filtered out and do not contribute to the final spectrum. The spectral window is set to 20 ppm for urine. In total 32 transients are acquired with 64 k data points for urine. The receiver gain was set to 90.5 for all of the experiments. 1H pulses were calculated in automation for each sample while the centre of the window was set on the water signal of the SR sample and kept constant during the acquisition of the whole dataset.</p><p><br></p><p><strong>J-resolved sequence</strong></p><p>The Bruker J-resolved pulse sequence (<strong>jresgpprqf </strong>in Bruker database) takes the form –RD-90°-t1-180°-t1-ACQ, where t1 is an incremented time period, RD is the relaxation delay (2 s), t1 is the increment delay, 90° represents a 90° RF pulse while 180° is the 180° RF pulse, ACQ is the data acquisition period. The spectral window is set to 16.6 ppm for the direct dimension and 78 Hz for the indirect dimension. 2 scans are acquired over 40 increments in the indirect dimension. Continuous wave irradiation is applied at the water resonance frequency using 25 Hz RF strength during the relaxation delay RD. More information about the pulse sequences used can be found in the previously published paper<strong>[1]</strong>.&nbsp;</p><p><br></p><p>Processing of the spectra (including Fourier transformation, phasing, baseline correction and calibration) was done in automation using <strong>TopSpin 3.6</strong> (Bruker Coorporation, Germany). For the 1-dimensional NOESY-presat experiment, the line broadening was set to 0.3 Hz, a 0-filling by a factor of 2 is used to produce 132 K data points for processing, and the first-order phase correction set to 0.0. The data was then automatically phased using only 0-order phase correction, the TSP was calibrated to 0.0 Hz (0.0 ppm). The processing for the J-resolved employed 0-filling by a factor of 2 is included in the F2 dimension and the digital resolution is increased to 256 points in F1 by 0-filling. The limits for spectral processing are set to ± 20 Hz. The raw data is then Fourier transformed, tilted, symmetrized around the central horizontal (F2) axis and automatically baseline corrected in the F2 dimension. The TSP signal is then calibrated to 0.0 ppm in the F2 dimension and to 0.0 Hz in the F1 dimension. Data was subsequently QCed assessing line width, water suppression and base line. Any spectra not complying the required criteria was rerun. Samples were remixed inside their tubes before the new experiment was acquired.</p><p><br></p><p><strong>Ref:</strong></p><p><strong>[1] </strong>Dona AC, Jiménez B, Schäfer H, Humpfer E, Spraul M, Lewis MR, Pearce JT, Holmes E, Lindon JC, Nicholson JK. Precision high-throughput proton NMR spectroscopy of human urine, serum, and plasma for large-scale metabolic phenotyping. Anal Chem. 2014 Oct 7;86(19):9887-94. doi:10.1021/ac5025039. Epub 2014 Sep 16. PMID:25180432.</p>
Study Protocol URI									
Study Protocol Version									
Study Protocol Parameters Name		Post Extraction;Derivatization	Chromatography Instrument;Autosampler model;Column model;Column type;Guard column	Scan polarity;Scan m/z range;Instrument;Ion source;Mass analyzer			NMR tube type;Solvent;Sample pH;Temperature	Instrument;NMR Probe;Number of transients;Pulse sequence name;Magnetic field strength	
Study Protocol Parameters Name Term Accession Number		;	;;;;	;;;;			;;;	;;;;	
Study Protocol Parameters Name Term Source REF		;	;;;;	;;;;			;;;	;;;;	
Study Protocol Components Name									
Study Protocol Components Type									
Study Protocol Components Type Term Accession Number									
Study Protocol Components Type Term Source REF									
STUDY CONTACTS
Study Person Last Name	Correia
Study Person First Name	Gonçalo
Study Person Mid Initials	
Study Person Email	gd2212@imperial.ac.uk
Study Person Phone	
Study Person Fax	
Study Person Address	
Study Person Affiliation	Imperial College London
Study Person Roles	submitter
Study Person Roles Term Accession Number	http://purl.allotrope.org/ontologies/role#AFRL_0000441
Study Person Roles Term Source REF	roleAFRL
