Home

file - BioMed Central

1. vuM g e T Example Library MSRI Microsoft Excel EPIIT SE Home Insert Page Layout Formulas Data Review View Add Ins x A Calibri ha 0 wii a f Conditional Formatting 9 Insert E AF d Ga dfi Format as Table 3 Delete g pas g Bzu BH o A ms wissis s j8 28 Cell Styles Format 7 2 GC Clipboard Font E Alignment E Number E Styles Cells Editing E15 Dis f A C D 1 Name RI Quantifier lons ID Mass Spectrum 2 Imaginary Metabolite 1 2TMS 994 102 1 42 9 43 18 44 15 45 49 46 4 47 7 56 6 57 7 58 68 59 45 6 3 Imaginary Metabolite 2 MX1 3TMS 1000 4 152 163 2 40 2 41 2 42 3 43 19 44 4 45 22 46 2 47 4 49 1 50 3 51 18 4 Imaginary Metabolite 2 MX2 3TMS 1010 2 152 163 3 40 2 41 2 42 3 4319 44 4 45 22 462 47 4 49 1 503 51 18 5 Makebelievic acid 3TMS 1019 174 147 85 40 9 41 8 42 42 43 71 44 24 45 133 46 11 47 19 53 6 54 7 6 Nonsensamine 2TMS 1022 8 234 5 86 20 119 38 121 10 133 95 146 288 154 32 160 12 205 23 K Example Library Sheet2 Sheet J EM ob u 1 Ready PEE l i The order of the columns is not important as long as the column labels are exactly as shown The data to put in each column is as follows Name his is the name of the metabolite derivative It does not have to be unique to a particular entry so having several entries with the name Alanine 2TMS for example would be fi
2. Umconfsmed 1207 RH849 4 MZ285 28 906 1848 4 289 D Fructose derivative 2 ID211 RI1867 7 MZ364 29 2861869 4364 BTE D Galactose methoxime 5TM5 ID216 RI1876 3 MZ160 29 425 1877 160 DET Mannitol 6TMS 1D222 RI1911 8 MZ319 30 038 1911 2319 DD L Lysine TMS ID223 RI1912 1 MZ156 30 0511911 9 156 DE L Tyrosine 3TMS 1D228 RI1933 MZ218 E 30 3711932 6 21 sf L Tyrosine 3TMS 1D228 RI1933 MZ280 30 37 1932 7 280 IDE D Fructose 6 O 2 5 4 6 tetrakis O trimethylsilyl D glucopyranosyl 1 5 4 5 tetrakis O trimethylsilyl 50 587 1946 6 361 Now wait a moment and you will be automatically taken to the relevant chromatographic overlay in the Haw Data Viewer like this Chromatographic View 3 De A XIC Overlay RT 29 951 to 30 151 min mz 156 i 050407 ANTIA3H MEOH 3H 1 050407_ANTIA3H MEOH 3H 2 e 050407 ANTIA 3H MEOH 3H 3 050407 ANTIA 3H MEOH 3H 4 2 00e 050407 ANTIA 3H MEOH 3H 5 050407 METHANOL 3H 1 e 4TMS Intensity 050407 METHANOL 3H 2 050407 METHANOL 3H 3 050407 METHANOL 3H 1 00e ANTA MAUS CDU CC MM 0 006 04 Z i 29 9 30 30 2 Retention Time min This makes it really easy to compare heatmap values with the raw signals and peak detections that underlie them The Matrix Explorer loads any data matrix into an interactive heatmap in the same way as illustrated above without carrying out any renormalisation so now let s move on to the Comparativ
3. 030407_ANTIA 0H MEOH OH 1 XIC 030407_ANTIA 0H MEOH OH 1 XIC 030407_ANTIAQH MEOH OH 2 XIC O30407_ANTIAQH MEOH 0H A AN 030407 ANTIA 0H MEOH OH 3 XIC 030407 ANTIA 0H MEOH OH 3 XIC 030407 ANTIA 0H MEOH 0H A 030407 ANTIA 0H MEOH OH A 030407 ANTIA 0H MEOH 0H 5 XIC 030407 ANTIA 0H MEOH OH 5 XIC Label Statistically Significant L Min Fold Change Max p value Display Library Match Results Please Select a Match Report for Peak Annotation GCMS ID Batch Report_10 September 2008_4 19 23 MATCHREPORT Once loaded select which run to show peak annotations for Display 3 4 The raw data viewer control panel To load some raw GC MS data into the viewer you must first select some file s using the Raw Data Viewer Control Panel This is probably a good time to explain how the control panel works Here s a close up Select chromatograms for p lotting in blue here Select chromatograms for plotting in red here Raw Data Viewer Control Panel Red Chromatograms ee Oe A 030407 ANTIA OH MEOH OH1XIC A 030407 ANTIA DH MEOH 0H 2 XIC 030407_ANTIA OH MEOH OH 2 XIC 030407_ANTIA 0H MEOH OH 3 XIC 1 030407 ANTIA OH MEOH OH 3 XIC E 030407 ANTIA DH MEOH 0H 4 XIC 030407 ANTIA OH MEOH OH 4 XIC 030407 ANTIA OH MEOH 0H 5 XIC A 030407 ANTIA 0H MEOH OH 5 XIC m z TIC 0 147 q Enter the m z of interest here entering 0 gives you the Total lon Chromatogram available Bl Label Statis
4. Import Network from Table Data Sources Input File Suspension Antimycin A Timecourse GCMS ID Batch Report_11 August 2008_5 14 47_CYTOSCAPE_CORNET_19 August 2008_1 9 58 TXT Select File s Interaction Definition Source Interaction Interaction Type vi Column 2 Target Interaction CME Gy Columns in BLUE will be loaded as EDGE ATTRIBUTES Column 1 Advanced Show Text File Import Options Text File Import Options Delimiter Tab Comma Semicolon Space _ Other Preview Options Show all entries in the file 8 Show first entries Network Import Options Default Interaction pp Attribute Names C Transfer first line as attribute names Start Import Row 1 Comment Line Preview P Text File Left Click Enable Disable Column Right Click Edit Column GCMS_ID Batch Report_11 August 2008_5 14 47_CYTOSCAPE_CORNET_19 August 2008_1 9 58 TXT e Column 3 e Column 4 e Column 5 I Trimethylsiloxy 2 trimethylsilylaminoe 1 Positive Glycine 2TMS ID18 RI1106 4 MZ204 Glycine 2TMS ID18 RI1106 4 MZ102 Positive Pyridine 3 trimethylsiloxy MS80 RI10 Positive Phosphonic acid methyl bis trimethyl Positive n dodecane ID29 RI1200 MZ41 Positive n dodecane ID29 RI1200 MZ57 Positive n dodecane ID29 RI1200 MZ71 Positive n pentadecane ID105 RIT500 MZ57 Positive e Column 1 e Column 2 I Trimethylsiloxy 2 trimethylsilylaminoe I
5. Value 7683B Automatic Liquid Sampler Agilent Technologies Palo Alto CA USA Samples were injected into the split splitless injector operating in splitless mode with an injection volume of 1 uL purge flow of 50 mL min purge FrmEmO n kenn md CPEB STANDARD MSRLMSRI Microsoft Excel Sien e pv Conditional Format as Cell Formatting Table Styles Value 7683B Automatic Liquid Sampler Agilent Technologies Palo Alto CA USA Samples were injected into the split splitless injector operating in splitless mode with an injection volume of 1 uL purge flow of 50 mL min purge time of 1 min and a constant inlet temperature of 300 degrees C Agilent 6890N Gas Chromatograph Agilent Technologies Palo Alto CA USA Agilent GC MSD Productivity ChemStation Software ver D 02 00 SP1 Agilent Technologies Palo Alto CA USA Product No G1701DA 30 m long 0 25 mm internal diameter Varian FactorFour VF 5ms capillary column with 0 25 um thick 9596 dimethyl 596 diphenyl polysiloxane film and 10m integrated guard column Varian Inc Palo Alto CA USA Product No CP9013 99 999996 pure helium with built in purifier was used at a constant flow of 1ml min Oven temperature was kept constant at 70 degrees C for 1 min ramped to 76 degrees C at 1 degrees C min and then ramped to 325 degrees C at 6 degrees C min and held for 10 min Agilent 5975B Inert MSD quadrupole MS detector Agilent Technologies Palo Alto CA USA The trans
6. An identical copy of the plot is also provided in Scalable Vector Graphics SVG format for editing in a compatible drawing program like Adobe Illustrator Below these plots you will see a scree plot 20 40 30 20 Percent of Total Variance 96 DA PCS PCB PCT PC amp PC9 PC10 Principal Component Below the scree plot you will see a variable loadings heatmap You can sort the columns by clicking on the headers Here we have sorted by the most biologically interesting component PC2 You will find that the metabolites with the highest absolute loading values in principal components that give good separation between two biological classes will also be the most significantly different ones when a t test is used to compare those classes VARIABLE LOADINGS Hinks Click an tha PC column headers to sort table contents PCI PLC 2 PC 4 PC4 PCS PCR PC PCR PCS PCW Unknown E1335 8 Ui RSS 8 MES i Unknown R23 1 2 ID38 F11231 2 M7234 Ajab acid TIMES 1055 RIET3OG 3 Meo Ghyceric acid 3TM amp 7 51330 8 MZ297 L Valin GTM IDSs IZ A Mids Aminu ylopropanecarbonh ip acid TAS IDO3 Ian 4 M20 1ummocyclopopanctaboryliec acd TMS Das ba T M20 Sedoheptul meihoxime ETMS M536 RI2115 SLIDES RI2115 5 7315 L Prob Z TMS ID RI 3200 Mid degen Ri2T57 2 OI RIZIST 2 MZ361 DF nactose 5 phosphate methaxime BTE EZ Peak O65 RO 9 Mz115 beta Alani ine STS IDES RHI1428 MIMB Laie acid ATMS ED Atia HB Mza L a me 2ZTM5
7. Biclustering of Metabolite X Sample Data Matrix v Remove redundant analyte signals Remove metabolites of unknown chemical structure NOTE Internal Standards and known Artifacts will be removed from the data set automatically Variables will be scaled to unit variance in Metabolite x Sample Biclustering Then select the 24 hour Antimycin A and Methanol treated samples from the selection box and hit Submit Sample Selection Please select the samples to be included in the HCA 060407 ANTIA 24H MEOH 1 PEAKLIST 060407 ANTIA 24H MEOH 2 PEAKLIST 060407 ANTIA 24H MEOH 3 PEAKLIST 060407_ ANTIA 24H MEOH A PEAKLIST 060407 ANTIA 24H MEOH 5 PEAKLIST 060407 METHANOL 24H 1 PEAKLIST 060407 METHANOL 24H 2 PEAKLIST 060407 METHANOL 24H 3 PEAKLIST 060407 METHANOL 24H 4 PEAKLIST 060407 METHANOL 24H 5 PEAKLIST Wait a moment and if you have your PDF viewer plugin installed correctly you will be presented with a heatmap clustering like this Ded3p ANTIA 24H MEOH A PEARL IST D ANTIA 24H WEGH 5 FEAKLIST DegaD ANTIA 23H MEOH 2 FEAKLIST D ANTIA 24H MEOH 1 PEARILIST D ANTIA 24H MEOH 3 PEARTLIST WHDI METHANOL 24H S PEAKLIST Ded4pz METHANOL 25H 4 PEAKLIST Deddp METHANOL 22H 3 PEAKLIST D METHANOL 25H 2 PEAKLIST Ded4pz METHAMOL 25H 1 PEAKLIST If you choose to do biclustering on the Metabolite x Metabolite Correlation Matrix you will get a result like this Micotinase argine eun die feanine
8. Step 3 Perform Data Import and Peak UDetechon 7 step 4 Upload sample information table ccccccsececssceceeeceeeeceueeseeseseueessaees 7 step 5 Upload a retention index calibration file 10 1 4 2 Choosing a research area and importing sample information 11 1 5 Building and using custom Mass Spectral and Retention Index MSRI Jie gs RIS MINNIE 19 1 5 1 An overview of the MetabolomeExpress MSHI library format 13 1 5 2 Creating MSRI libraries from AMDIS MSL files 15 1 5 3 Adding metadata to MSRI libraries eeeeseeeseesssesse 18 1 5 4 Displaying and Validating MSRI Libraries with MSRI Library Manager 20 1 5 5 Using Analyte Annotations Tables to customise data filtering 20 1 6 Supported Data Formats including example files 22 1 6 1 The main MetabolomeExpress metadata exchange format C METADATA TA EEN 22 1 6 2 Raw GC MS data NetCDF and the MetabolomeExpress eXtracted lon Bergerie acta nie O Eo EE 27 BC POG NSE TAI PEARLS T EE 27 1 6 4 MSRI library matching report tables MATCHREPORT 28 1 6 5 Data matrices ER KEE 29 2 He analysis of publicly disseminated data using Database Explorer 31 2 1 Getting an overview of the database with the Database Statistics
9. and one giving some breakdown information on each of the experiments associated with each of the publications represented in the database Each publication may be associated with a number of experiments investigating different hypotheses You can link out to articles in PubMed by clicking on the PubMed hyperlinks You may also load the dataset into the Experiment Explorer module by clicking on the little green flask icon Ai next to the experiment name RA Home A Experiment Explorer lig Database Explorer ie Help 1 amp Registration Database Statistics ResponseFinder MetaAnalyser MSRI Library Manager Database Statistics Display Current Contents of the MetabolomeExpress Peer Reviewed Metabolomics Database Summary Statistics Attribute Value No of peer reviewed publications 8 No of experiments 11 No of metabolome comparisons 70 No of metabolite response statistics 9772 No of species 3 N No of genotypes 13 No of tissue organ fluid types Publication Information NOTE Click A to load an experiment into Experiment Explorer 7 Publication Pubmed i S GIE 2x p pa d SC Number of Publication Reference Experiment Name Genotype s Organ s Treatment s Metabolites Giraud E Ho LH Clifton R Carroll A Estavillo G Tan YF Howell KA Ivanova A Pogson BJ Millar AH Whelan J 2008 The Absence of Combined Moderate High Light and Drought Stress ALTERNATIVE O
10. ANALYTICAL SAMPLE PREPARATION PROTOCOLS Analytical Sample Preparation Pro Analytical Sample Preparation Protocol ID Analytical Samale Preparamon Protocol Descrmtian Analytical Sample Preparation Protocol 1 BIOSAMPLES BioSample 1 Brame ID Explant 044 otkaemnple Precesung Molota ID BonSaenple Mass Volume Biosaenple Mass Velum Units Preparatian Date Pree MM IR BioSample 2 Extract 1 l a tract En Sanbarme e mi Extraction Prosecco ID Preparation Gate ev MA DD Extract 2 ANALYTICAL SAMPLES Analytical Sample 1 Analytical Sample IO Extract iD Analytical Sample Preparation Pratocal ID Preparation Date mr oh CHT Analytical Same INSTRUME L PARAMETERS Instrumental Parameters Au cernpector Chron eg phy biir ueni risbrumaeit Coni ol 5oftwarne Seperation Colurrin Separation Parameter Aiii Spec Enorme try brumen BATCH SEQUENCES Batch Sequence 1 Sabon Sequence H3 vr bhurpbaemt ba Dapp Stared nne Aa Dk RI Calibration F e Batch Sequence 2 ANALYTICAL RUNS Analytical Run 1 Sun RX Nett Ce hlenueme wihout LS expenmsian Anabytscal Sample ID Aun Dui Time OW Y MAM DD Hun NERAD Ne peguea W orma sn NetCOS file Batch Segoesnor ID Analytical Run 1 NOTE The structure shown above was designed for plant metabolomics MetabolomeExpress support customised field sets and validation schemas for 18 different research areas To make sure your
11. Datasets are submitted using the HesponseFinder MySQL DB Stats Import tab of the Statistics and Data Exploration Control Panel see below Statistics and Data Exploration Control Panel e Match Report to Data Matrix Data Matrix Re nalisatior Matrix Explor Comparative Statistics PCA HCA Correlation Analysis ReponseFinder MySQL DB Stats Import Please select a field of research for selection of appropriate metadata validation M Plant Arabidopsis thaliana se select a STATS file for import in Era ID Batch Report 2010 4 6 ee aca 2010 geg e NONREDUNDANT STATS Refresh To submit your dataset simply select the field of research corresponding to your metadata file version and the STATS file containing the statistical results you wish to submit and then click the Import button Your metadata file will then be checked against the appropriate template and if it passes validation the MetabolomeExpress curator will be emailed with a notification of your wish to submit the dataset A final manual security check will be conducted and if passed the dataset will soon be imported into the database You will be notified by email as soon as this is done If your dataset does not pass validation check the results panel for error messages indicating which fields are causing failure and fix your metadata file or required data files accordingly before trying again 7 APPENDIX A Interpretation of MetabolomeExpr
12. SQ In the above screenshot the response of suspension cultured Arabidopsis cells to 16 hours of rotenone treatment inhibition of mitochondrial respiratory chain complex has been selected as bait Clicking the GO button yields the following results t Home A ieee Explorer E Database Explorer o Help X Registration Database Statistics CURES pon PhenoMeter MSRI Library Manager PhenoMeter Display PhenoMeter Control Panel v Um a am J 7X Publication Garmier M Carroll AJ Delannoy E Vallet C Day DA Small ID Millar AH 2008 Complex I dysfunction redirects cellular and mitochondrial metabolism in Arabidopsis Plant Physiology 148 3 1324 41 Experiment Timecourse Treatment of Arabidopsis cell suspension cultures with rotenone Class Comparison At Ler Cell Suspension Cell suspension Rotenone 16h At Ler Cell Suspension Cell suspension Methanol 16h E Publication Experiment Class Comparison ett 3 Garmier M Carroll AJ Delannoy E Vallet C Day DA Small ID Millar AH 2008 Complex I dysfunction redirects cellular and Timecourse Treatment of Arabidopsis cell suspension cultures with At Ler Cell Suspension Cell suspension Rotenone 16h At mitochondrial metabolism in Arabidopsis Plant Physiology 148 3 1324 41 rotenone Ler Cell Suspension Cell suspension Methanol 16h Garmier M Carroll AJ Delannoy E Vallet C Day DA Small ID Millar AH 2008 Complex I dysfunction redirects
13. left click Generate MetabolomeExpress MSRI Format The screenshot below shows an MSL format version of the Q_MSRI library named mpimp msl from the Max Planck Institute for Molecular Plant Physiology MPIMP Golm Metabolome Database being selected for conversion to MSRI format Mozilla Firefox Fle Edt Wew History Bookmarks Took Help Q9 c x o e ee E SH nitrogen generator kims ZC BM L Most Visited Pin btps wweanerfEdesphp O OOOO mu 8 metabolomexpress org Wavigation di Home Geet Explorer Mg Database Explorer i ue 2 Reoscracc Database Mav D anase Sones R spnpzetp er Methane tar MSA Library Manager j Plant norgyBictogy Internal e MSHI Library a C Fiani nergyBietegy Putte Manager Display a Danu RSB ME Fact Pubit a AMU REB MS Fact inbernal H5RI Library Hangcer Control Parel R ght cikk on a library to access library management oppor Lp PlantErergsBiology Public HERI Libraries D a ij Plantinergyfictogy in house MSA Libraries 37 Ade dih Cord aness Unites mal map T OPED STANDARD SRI LSA EJ CFEB STANDARD MSRI Ribilal Ri 333 MS 3j CPEB ETAHDARD USRI Ribilot Ri 1722 ACH 1a COLD 48h MERH T OPES STANDARD WSR Ribitel Ri 1723 HP 1 FIEHN RECONSTRUCTION MARI A FEHN RECONSTRUCTION 3 MSR 2 Fab Annatason MSRI S STANDARD MERI_010307_curated_quantons new standards al8 unknows combimis standards and c 3 mpimp mat man a E n hydracam on mast
14. nikimate prose 1 hylalanine ipa E SC Be GE EH Eu Lama 7 t ger ie ues pne Ge LGE P S SE St rate De de me E Fetal Sdt Z n Gros snosine 5 monophosphate Ahane uere gj acras mpanecarbenzte true 8 Bee n CR Srpxyproune j ene i AMINE BOTEN maly s Alsnine ste gt sence acid 3 phosphate oe hospha Smaspnate c Bas ise phosphate kee LZ ee eee H HSH Ei EI mug Li d pra a ica mi uc eu t LEE c1 HoH UCHR DENM S EE BE Lu exe C ELO CR MRS COME BT aa I d d A E c x m 3 z gu GE BONUS ES GUERRE oe DE ORA PE d e Pos BTE LR P 3 5 SS ag d R wi g a i d Qa E e ES EO E Zz 2 S 2 e Kl i Es d E 6 7 Correlation network construction In addition to doing hierarchical bi clustering of metabolite metabolite correlation matrices MetabolomeExpress also has a tool for generating correlation network graphs from data matrices The tool is very simple All you need to do is select a data matrix from the drop down list specify a minimum correlation coefficient that two metabolites must exhibit in order for their nodes to be connected and hit the Upload button The selected parameters will then be sent to the server and your network graphs will be constructed using all the values in the data matrix For the purposes of example try setting the Correlation Analysis Control Panel up like this and hit Upload Statistics and Data Exploration Control Panel Match Report to Data M
15. the thresholds it is probably not a real peak and will be discarded 4 2 The Peak Detection Control Panel The figure below explains what all the different parts of the control panel are for P This is the critical slope value that an EIC signal must exceed QUINT ONCOL RITIRO IM NND Peak Detection Settings in order to alert the peak finder algorithm that it may be reading 030407 ANTIAOH MEOHOH IXIC A Shope threshold 50 into the start of a chromatographic peak esi eid tan ttn Mire i A This is the minimum integrated peak area for a potential peak to 030407 ANTIA 0H MECH DH 3 XIC Mn Pekha 500 E 030407 ANTIA OH MEOH DH 4 XIC Mus ER considered a real peak 030407 ANTIA OH MEOH DH 5 XIC PNE hy isi Kean inht fin arhi EE See LS enen 030407 METHAHOL OH A XC Min oe 15 rei 030407 METHANOL DH 3XIC mins The peak purity factor is defined as 030407_METHANOL OH 4 IC J IER a Total integrated signal Integrated signal lying below the lowest integration point GARE ecu A Where the lowest integration point is the signal intensity at either the peak start or peak end whichever is lower e This is the minimum peak width in scans for a potential peak to be considered a real peak The peak width is measured from the integration start point to the integration end point Check this box if you want to extract peaks from the selected files Otherwise netCDF files will be used to generate XIC files but no PEAKL
16. 72 131 D Glucuronic acid 2 3 4 5 tetrakis O trimethylsih trimethylsih l ester MS79 Unknown 0 74 1 36 RI2367 2 Galactinol STMS MS84 Unknown 0 76 1 32 Nicotinic acid Amino Acids 0 85 1 51 Amina If you see an interesting result like the 162 fold increase in the unknown metabolite with MS similarity to uric acid you can double click the cell containing its signal intensity ratio and be taken to the raw GC MS signals in the raw data viewer Lal Home A Experiment Explorer TP Database Explorer o Help amp Registration i e M ee MetaData Viewer Raw Data Viewer Data Import and Peak Detection MSRI Library Matching Statistics and Data Exploration Chromatographic View Resize Selection left click gt move mouse gt left click Zoom in Hold A double click Raw Data Viewer Control Panel v Zoom out Hold Z double click d View spectrum Hold Shift double click EIC Overlay RT 32 762 to 32 962 min mz 441 Uric acid 4TMS MS81 RI2094 8 Putative but Unconfirmed ID251 RI2094 8 MZ441 6 0e 5 5 0e 5 4 0e 5 Intensity 3 0e 5 2 0e 5 1 0e 5 E 0 0e 1 Retention Time min You can copy and paste the results into Excel for offline analysis if you like You can also download a 2D HCA clustergram in PDF format using the provided hyperlink Here is a thumbnail image of the clustergram from this example analysis mm Tug Sab pe Feit 9
17. A Howell KA Carroll A Howell KA Carroll A Narsai R Howell KA Carroll A Rennenberg H Millar AH Rennenberg H Millar AH Rennenberg H Millar AH Ivanova A Millar AH Whelan J Whelan J 2009 Differential Whelan J 2009 Differential Whelan J 2009 Differential 2009 Defining core metabolic response of gray poplar response of gray poplar response of gray poplar and transcriptomic responses to leaves and roots underpins leaves and roots underpins leaves and roots underpins oxygen availability in rice stress adaptation during stress adaptation during stress adaptation during embryos and young seedlings hypoxia Plant Physiology hypoxia Plant Physiology hypoxia Plant Physiology Plant Physiology Epub ahead of Chemical Average Standard EE Ewe Ee e Metabolite Name class Responsiveness Deviation 149 1 461 73 149 1 461 73 149 1 461 73 print Timecourse Flooding of Timecourse Flooding of Timecourse Flooding of Rice germination timecourse Poplar roots Poplar roots Poplar roots Air N2 gas switch Populus x canescens Root Populus x canescens Root Populus x canescens Root Rice WT seedling Flooding_24h Populus x Flooding_5h Populus x Flooding_168h Populus x Air_N2_Switch_48h Rice WT canescens Root canescens Root canescens Root seedling Air_48h Control_24h Control_5h Control_168h Uac ada dins a Putative known 58 63 72 74 Unknown Similar to Proline 2TMS RI1614 8 Unknown 0 68 121 DEES Unknown 0
18. DH 1 IC a8 Q30407 ANTIA OH MEOH OH 2 COF E 030407 AHTIA DH MEOH OM Z PEAKLIS 37 030407 ANTIA OH MEOH OH 3 ME 030407 AHTIA GH MEOH DH 3 CDF T 030207 ANTIA H MECH DU 3PEAKLIE Z ii Q30407 ANTIA OH MEOH OH 3 CC 030407 ANTIA OH MEOH DH A CO 37 30407 ANTIA OH MEOH OH 4 PEAKLIE z 0204D7 AHTIA 4 MEOH DH 423c ii Si 8 30407 ANTIA OH MEOH OH COF 030407 ANTIA OH MEOH DH 5 PEAKLIS E 3 s07 AMTIA OH MEOH DH 5c 3 30407 METHANOL OH 1 CDF 3 Si c 020407 METHAHOL G4 1 PESKLIST 030207 METHANOL OH 1 XIC 030407 METHANOL DH 2 COF E 030207 METHANOL OH 2 PEAKLIET Z DI 030407 METHANOL OH 7 XIC 030407 METHANOL OH 3 CC T 030207 METHAHOL GH 3 PEAELIST amp 030407 METHAHOL DH LOC E 3 s07 METHANOL OH 4 CDE r Ferenn Contact Information Lkenseg x Find rte m 130407 AHTIA GH ME DH OH T PIEAKLIS B bes tt Previous Heghligha al A Lapwargmant Explorer Example Experiment Arabsbopesis Cel Antnin A Timecoure iu Dabo Explorer o Hep A Regetraten amp tatsstics and Data Exploration d Horne Matalata Viewer Aare Data Viewer Data Import and Paak Detection MERI Library Matching User authormaben OK MetaData Ge found metadata Tesue masses volumes vill be assumed to be equal and no dass annatabons wi be appbed ro sample columns Output TA MATRICES The folowing data matrices have been ge
19. Hypoxia cultured cell 1 100 7 3151009 wildtype hypoxia Rep6 wildtype cell line Hypoxia cultured cell 1 100 8 151009 wildtype normoxia Rep 1 wildtype cell line Normoxia cultured cell 1 100 9 151009 wildtype normoxia Rep2 wildtype cell line Normoxia cultured cell 1 100 210 151009_wildtype_normoxia_Rep3 wildtype cell line Normoxia cultured cell 1 100 11 151009_wildtype_normoxia_Rep4 wildtype cell line Normoxia cultured cell 1 100 12 151009_wildtype_normoxia_Rep5 wildtype cell line Normoxia cultured cell 1 100 13 151009_wildtype_normoxia_Rep6 wildtype cell line Normoxia cultured cell 1 100 14 151009_cancer cells_hypoxia_Rep1 cancer cell line Hypoxia cultured cell 1 100 15 151009 cancer cells hypoxia Rep2 cancer cell line Hypoxia cultured cell I 100 16 151009 cancer cells hypoxia Rep3 cancer cell line Hypoxia cultured cell 1 100 17 151009 cancer cells hypoxia Rep 4 cancer cell line Hypoxia cultured cell 1 100 18 151009 cancercells hypoxia Rep 5 cancer cell line Hypoxia cultured cell 1 100 19 151009 cancer cells hypoxia Rep6 cancer cell line Hypoxia cultured cell 1 100 20 151009 cancer cells normoxia Rep 1 cancer cell line Normoxia cultured cell 1 100 21 151009 cancer cells normoxia Rep 2 cancer cell line Normoxia cultured cell 1 100 22 151009 cancer cells normoxia Rep3 cancer cell line Normoxia cultured cell 1 100 23 151009 cancer cells normoxia Rep 4 cancer cell line Normoxia cultured cell 1 100 24 151009 cancer cells normoxia Rep5 canc
20. IDAT Rizr MEISS L Alanine Hia ihya ester h s FT 6 ICI RD amp M24 unio n RITZ217 2 ID36 FR 21 r2 METAT L Tyrasimea XTMS ID278 RSIS METH L nom ina OI E IOF RBI1355 8 MUST p Amin rie acad 2TM S IDS Riq838 1 MI 781 L4 Phim laning ex ep ID RI ke pr Pyrog Amamuc acad TMS ID115 RHSZ1 2 ME156 Arakenoturanose 1 2 3 ebak a D mmpgplbreisdhel kt MS mmer 5 1143 RiT607 5 M Uniknya HRZ 2 IDZ35 GE 2 Men Unknown Possible Organic Acid RI1411 1 Dra Rd T M19 H T irmethyteihyl 2 pyrrolidmone KESSE BITTE WIE FRITT3 1 a Ba D KArabinose m theoxime TMS ID1561 RI1559 6 N17 Z mmer acid 3TMS ID1 MO RT MEZSU Unknawn R 2731 2 Su RO 3 Mz425 L fryptophan GIN 261 RE2210 3 ae Raffinase 117M Maa Putative but Unconfinmed ID342 RI3341 3 MEAT L Tyrosime 3TMS ID228 RI1833 MZ2SU 4 Keteglucose bis O methyloxime tetrakis trimathylsihjT MSC RHEL ID33 RI1463 6 METT3 ET A jJ ET Finally if you have an XSD compatible virtual reality browser plug in installed you will see a 3D PCA Score Plot showing the first three PCs Here s a screenshot showing what the example PCA looked like when viewed using the trial version of the BS Contact X3D viewer available for free from Bitmanagement Software If you like you can download the X3D files using the hyperlink at the top of the page and edit the 3D scene using an X3D compatible 3D editor to create a nice powerpoint slide for example 6 6 Hierar
21. MS files CDF for that experiment into the folder you just created If you don t have your files in NetCDF format you should be able to export them from your instrument manufacturer s data processing software HINT It will make things easier for everyone if you name your files descriptively rather than with meaningless numbers or letters Name your files something like 20091015 Nutritional Regime A Day 1 Replicate 1 CDF 20091015 Nutritional Regime A Day 1 Replicate 2 CDF etc where the first 8 numbers represent the date on which the batch sequence of GC MS runs began in YYYYMMDD format rather than something like 1 CDF 2 CDF etc Step 3 Perform Data Import and Peak Detection Now log in to MetabolomeExpress and you should be able to find your new experiment folder containing your data files in the Database Navigation panel on the left hand side of the interface To begin processing your files right click on the experiment folder name and left click Load folder contents Your files should now be ready for selection in the Data Import and Peak Detection Control Panel inside the Data Import and Peak Detection tab see Section 3 4 Data Import and Peak Detection You may perform the data import and peak detection process while completing the next two steps required for library matching data matrix construction and statistical analysis Step 4 Upload sample information table The next step is to provide basic sample inform
22. Whelan J 2009 Defining core metabolic and transcriptomic responses to oxygen S 5 E Rice WT seedling N2 Air Switch 27h Rice WT seedling availability in rice embryos and young seedlings Plant Physiology Epub ahead of print Rice germination timecourse N2 Air gas switch N2 27h y Narsai R Howell KA Carroll A Ivanova A Millar AH Whelan J 2009 Defining core metabolic and transcriptomic responses to oxygen se tan ae T Rice WT seedling N2 ir itch 30h Rice WT seedling availability in rice embryos and young seedlings Plant Physiology Epub ahead of print Rice germination timecourse N2 Air gas switch 2_30h Narsai R Howell KA Carroll A Ivanova A Millar AH Whelan J 2009 Defining core metabolic and transcriptomic responses to oxygen S S Rice WT seedling Air N2 oe s ch 48h Rice WT seedling availability in rice embryos and young seedlings Plant Physiology Epub ahead of print Josie an chet aoc eene Air 48h Howell KA Narsai R Carroll A Ivanova A Lohse M Usadel B Millar AH Whelan J 2009 Mapping metabolic and transcript temporal MEE Wes A A switches during germination in rice highlights specific transcription factors and the role of RNA instability in the germination process Rice aerobic germination timecourse Rice WT seedling Germ Air 24h Rice WT 15 din t Dry Seed 0h Plant Physiology 149 2 961 80 Narsai R Howell KA Carroll A Ivanova A Millar AH Whelan J 2009 D
23. Wildtype 2 Days on Diet A blood 2 100 15 151009 Wildtype Diet A Day 2 Rep 2 Wildtype 2 Days on Diet A blood 2 100 16 151009 Wildtype Diet A Day 2_Rep 3 Wildtype 2 Days on Diet A blood 2 100 17 151009 Mutant Diet A Day 2 Rep1 Mutanti23x 2 Days on Diet A blood 2 100 18 151009 Mutant Diet A Day 2 Rep2 Mutant123x 2 Days on Diet A blood 2 100 19 151009 Mutant Diet A Day 2 Rep3 Mutant123x 2 Days on Diet A blood 2 100 20 151009 Wildtype Diet B Day 2 Rep 1 Wildtype 2 Days on Diet B blood 2 100 21 151009 Wildtype Diet B Day 2_Rep 2 Wildtype 2 Days on Diet B blood 2 100 22 151009 Wildtype Diet B Day 2 Rep 3 Wildtype 2 Days on Diet B blood 2 100 23 151009 Mutant Diet B Day 2 Rep1 Mutanti23x 2 Days on Diet B blood 2 100 E 151009 Mutant Diet B Day 2 Rep2 Mutanti23x 2 Days on Diet B blood 2 100 25 151009 Mutant Diet B Day 2 Rep3 Mutanti23x 2 Days on Diet B blood 2 100 26 In some types of experiments such as human clinical metabolomics experiments different disease states are considered different treatments The example below shows a dummy clinical experiment investigating the urine metabolome for interactions between disease and drug treatment at two different time points with respect to some starting time point which you can explain later in the more detailed metadata format if necessary Al A B C D E Organ or Sample Sample ID Genotype Treatment Biomaterial Timepoint Mass or 1 Type Volume 2 151009 Wildtype Disease Drug Week 1 Re
24. after beginning of the diet feeding period Insert Page Layout Formulas Data Review View Add Ins Acrobat B cu Calibri elt IA a gt Sp Wrap Text General a i Bad Paste EC IB rg A mi kad Merge amp Center 9 gd 320 Conditional Format Format Painter I C rs z TT Formatting as Table Clipboard E Font Ta Alignment E Number E Styles M24 vO n Pl A B C D mur F 1 Sample ID Genotype Treatment Organ or Biomaterial Type Timepoint Sample Mass or Volume 2 151009 Wildtype Diet A Day 1 Rep 1 Wildtype 1 Day on Diet A blood 1 100 3 151009_Wildtype_Diet A_Day 1_Rep 2 Wildtype 1 Day on Diet A blood 1 100 4 151009 Wildtype Diet A Day 1 Rep 3 Wildtype 1 Day on Diet A blood i 100 5 151009 Mutant Diet A Day 1 Rep1 Mutant123x 1 Day on Diet A blood 1 100 6 151009 Mutant Diet A Day 1 Rep2 Mutant123x 1 Day on Diet A blood I 100 7 151009 Mutant Diet A Day 1 Rep3 Mutant123x 1 Day on Diet A blood 1 100 8 151009 Wildtype Diet B Day 1 Rep 1 Wildtype 1 Day on Diet B blood al 100 9 151009_Wildtype_Diet B_Day 1_Rep 2 Wildtype 1 Day on Diet B blood 1 100 10 151009 Wildtype Diet B Day 1_Rep 3 Wildtype 1 Day on Diet B blood 1 100 11 151009 Mutant Diet B Day 1 Rep1 Mutant123x 1 Day on Diet B blood 1 100 12 151009 Mutant Diet B Day 1 Rep2 Mutanti23x 1 Day on Diet B blood 1 100 13 151009 Mutant Diet B Day 1 Rep3 Mutant123x 1 Day on Diet B blood 1 100 14 151009 Wildtype Diet A Day 2 Rep 1
25. al c3 Garmier M Carroll AJ Delannay E Vallet C Day DA Small ID Millar A 2008 Complex dysfunction redirects c d el Timecaurse Treatment of Arabidopsis cell suspension cultures with rotenone z3l a Ler Cell Suspension Cell suspension Methanol 12h At Ler Cell Suspension Cell suspensio SO at Ler Cell Suspension Cell suspension Methanol_16h At Ler Cell Suspension Cell suspensio eelef Ler Cell Suspension Cell suspensian Methanol 1h At Ler Cell Suspension Cell suspension SE at Ler Cell Suspension Cell suspension Methanol 24h At Ler Cell Suspension Cell suspensio z3l at Ler Cell Suspension Cell suspension Methanal 3h At Ler Cell Suspension Cell suspension SO at Ler Cell Suspension Cell suspension Methanal_6h At Ler Cell Suspension Cell suspension eelef Ler Cell Suspension Cell suspension Rotenone 12h At Ler Cell Suspension Cell suspensic __ E at Ler Cell Suspension Cell suspension Rotenone 16h At Ler Cell Suspension Cell suspensic SC a Ler Cell Suspension Cell suspension Rotenone_ih At Ler Cell Suspension Cell suspensior SO a Ler Cell Suspension Cell suspension Rotenone_24h At Ler Cell Suspension Cell suspensic eelef Ler Cell Suspension Cell suspension Rotenone_3h At Ler Cell Suspension Cell suspensior EIC ar Ler Cell Suspension Cell suspension Rotenone_6h At Laer Cell Suspension Cell suspensior 4 Tm t Refresh
26. being displayed 1 Mozilla Firefox am hittps wwewam rg indexsphp e i metabolome xpress e um Expermant Explorer lig Database Explorer FT E Regitration Dube States Besonzefe der Heinizx icner H5SRI Library Manager MSR Library Manager Display mar S ien cT y 4 1047 Lactic acid 2TMS Lactic acid Matched this library E Mass Spectral Comparison nd stinni an Botiom l Library aa a a acid 2TMS _ RE ee CH lon specified TE 5 1054 Lac A Matched this library Gester E z be sal an Bottom ee No quantifier Library Spectrum RHOGT 7 L Alanine 2TMS _ KS ion specified on Top 6 10640 Dix jE3 Matched WIEN Dapley MS ih 73 entry will mot Bon H i NETT be matched LB Eh D 1B Wi r SEN No quantifier ae a Hn zn m An zn eu EA ipn specified 3 Gear Top 7 1064 788 Heraa cir rm foo Mekal cir this library Matched Display MS entry will not on Bol be matched uantifier 3 Display MS Not ion specified TES 8 1076 682 4 Aminobutyric acid 3TMS 682 4 Aminobutyric acid this bray p Matched Display MS entry will not on be matched Ki hips vw meraboieme ex press orq index papa EE E 7 Ve LI zeen E inboxin UH vw E sip Be Adobe D Eabb Rime Tomo 2 reef EU Opboard 1 5 3 Adding metadata to MSRI libraries It is possible to extend the basic MSRI library format in two ways One way Is to add additional columns of information to t
27. cellular and Timecourse Treatment of Arabidopsis cell suspension cultures with At Ler Cell Suspension Cell suspension Rotenone 12h At 36 mitochondrial metabolism in Arabidopsis Plant Physiology 148 3 1324 41 rotenone Ler Cell Suspension Cell suspension Methanol 12h Garmier M Carroll AJ Delannoy E Vallet C Day DA Small ID Millar AH 2008 Complex I dysfunction redirects cellular and Timecourse Treatment of Arabidopsis cell suspension cultures with At Ler Cell Suspension Cell suspension Rotenone 24h At 3 mitochondrial metabolism in Arabidopsis Plant Physiology 148 3 1324 41 rotenone Ler Cell Suspension Cell suspension Methanol 24h Garmier M Carroll AJ Delannoy E Vallet C Day DA Small ID Millar AH 2008 Complex I dysfunction redirects cellular and Timecourse Treatment of Arabidopsis cell suspension cultures with At Ler Cell Suspension Cell suspension Rotenone 6h At 19 mitochondrial metabolism in Arabidopsis Plant Physiology 148 3 1324 41 rotenone Ler Cell Suspension Cell suspension Methanol amp h Narsai R Howell KA Carroll A Ivanova A Millar AH Whelan J 2009 Defining core metabolic and transcriptomic responses to oxygen Rice germination timecourse germination in air vs germination in Rice WT seedling Germ N2 48h Rice WT seedling 18 availability in rice embryos and young seedlings Plant Physiology Epub ahead of print N2 Germ Air 48h Narsai R Howell KA Carroll A Ivanova A Millar AH
28. code examples below Code names for currently built in controlled vocabularies are described in the table below Type of Entity Name of Controlled Vocabulary Standard AGI convention names for genes of the model plant Arabidopsis thaliana eg AT1G48030 Obtained from ftp ftp arabidopsis org home tair Ontologies Gene Ontology ATH GO G OSLIM txt on 2010 04 01 Human gene symbols approved by the HUGO Gene Nomenclature human_genes Committee eg MDH1 Obtained from http www genenames org cgi bin hgnc_stats pl on 2010 04 01 Genes Mouse gene marker symbols as per the Mouse Genome Informatics MGI website eg Mdh1 Obtained from ftp ftp informatics jax org pub reports gene association mgi on 2010 04 01 Rice gene locus IDs as per the Rice Genome Annotation Project website eg LOC Os07g43700 Obtained from ftp ftp plantbiology msu edu pub data Eukaryotic Projects o sativa ann otation dbs pseudomolecules version 6 1 all dir all TU model brief info arabidopsis genes mouse genes rice genes Genes symbols for genes in the Drosophila melanogaster genome as per the FlyBase FB2010_03 release version of the file gene_association fb http flybase org static_pages downloads FB2010 03 go gene associati on fb gz Plant Ontology Consortium http www plantontology org terms for plant plant anatomy structure Obtained from http www obofoundry org cgi bin detail cgi id2po anatomy Fungal Anatomy Ontology Project te
29. lh MSRI T Library Manager 2 Oxoqlutarate Minimum Fold Change 2 fold Up OR Down v Maximum p value 0 05 Clicking GO retrieves the following results Greg Gepermentifepire Arabidapsis GT AHA SA and MOCO Dessen Mg Database xplorer nek Z Astrain Rams pesra ni miler aie WARI E my Maree ResponseFinder Search Results Besponsefinder Control Panel NOTE cick A to led an n experiment into Experiment Explorer E A t n STATS 4 Faki pale Publikan LE ELCH n US rali Dosage Dosase wle imaka Deechkan fatten F frkeregsclP IN waan Lo umesatceh ji aar Units ef fret ib tests Later R Wheian 1 Herve Millar A Probing the role ef fares ER 3009 The drabsidepon Ed guis ere trarsderase pene Demy o Oe Argent BHA PAg Calg reei i depuayscsmclex great reguaten E femtmentofa tote Adam J Caral Deche D NN haaa M Cai i hours x e m memet Up pos BC EE EE be KE Dreakenb METADATA DI feeding mak 24h S d tz iri itera me hate with ether eerily Fo ralem ore er G Plant Journal Epub sip P Carroll A RN Nene saprei E dag the rale af e 2609 The T za panne tratelerasa gane Fandy aey DEDO GF r Ger Btn EA EI E daer zerek eguaten SC enmen GA Fee Ades J Corral Dechen HMM sa 2h Colt u hus 1 milicie Gre DU RAE INN Uo and ce zlenora genes AT cl EE ie Treg ibs METADATA TIT needing mack 24h d juan LOTH TIE Dd ell ree ae t Plane ecl Epub ahead of p Garre
30. mari amp A 35 AO Ta 48h Cold Siers Unkrneowns rmsl T mpimp mal Relesn E n htrarart Display and Validate v E Gamma ina Tab dolum ed IERI Fontal Licensing Ot Sowa Code hittps www metabolome express org index phpa a Once a library is selected for conversion wait a few moments while the library is converted checked and displayed in the main window of the MSAI Library Manager as shown below Y Mozilla Firefox m x fle Edt Yew History Bookmarks Joos Help D I C X ve Ir EE npsJwww metabolome express org ndex php Cy A8 nitrogen generator ic ms ARA DI Log out uli Most Vieted metabolomexpress org Gore A eer Epiorer Mg Database Explorer G Heb uoa ptabare St C veer MSRI Library Manager MSRI Library Manager Deplay MSRI Library Format Conversion Results You Metabolometixoress Lab delmted SRI format Kar gr may be donniosded ver the irk below MSRI Library Contents Metabolite Library Spectrum R paar bali SE Analyte Name Metabolite Name Name Entry Lee D Matched ID x R Mass Spectrum No quantifier rate os Mass Spectral Comparison MSRI Library Manager Control Panel Not 1 1036 6 872 Pyruvic acid methoxyamine TMS 872 Pyruvic acid this library Matched entry will not Diphya be matched No quamer em Not jon specified SEHR 2 1038 Decamethyltetrasiloxane Decamethyltetrasiloxane this library Matched entry will not Oreley MS be matched No
31. of metabolite response statistics This will allow other researchers to use your data with the tools in Database Explorer To successfully submit your dataset your dataset will need to be in your public FTP repository folder and it will need to pass validation Validation is done automatically by a computer script The main purpose of validation is to check that all your raw and processed data files are present that your dataset has been adequately described and that you have used the correct ontologies and string formats for your organism and research area Therefore you will need to complete all the required fields in your METADATA TXT metadata file in accordance with the validation template file indicated for your area The template applicable to your dataset should appear in the VERSION section of the metadata file you created during sample import NOTE If you haven t seen your METADATA TXT file before check your experiment folder for a file with the name Name of Experiment Folder METADATA TXT See the section in this manual on supported file formats for information about these files You can download the current validation templates from the Help page on the MetabolomeExpress website You will need to refer to APPENDIX A in this manual to understand the templates You should also refer to any minimal metadata reporting guidelines available for your research area to make sure the metadata values you provide meet these guidelines
32. of the matched quantifier ion signal across the entire row of the table m z The m z of the quantifier ion used to quantify the analyte represented by this row The layout of the mzrtMATRIX format is shown in the screenshot below 3 WM 9 T GCMS ID Batch Report 10 September 2008 4 45 58 mzrtMATRIX Microsoft Excel Home Insert Page Layout Formulas Data Review View Add Ins A ms n Calibri e DL ftem le A is pnm 8 ER wrap Text General Y H E I ER E Sr d e EI CR Paste lp vw amp am A SIE Merge amp Center sl A8 Conditional Formatas Cell Insert Delete Format Sort amp Find amp z Formatting Table Styles X Y E Filter Select Clipboard Font E Alignment fa Number E Styles Cells m AK9 vO fe A B C D E F 1 Data File 030407_ANTIA OH MEOH OH 1 PEAKLIST 030407_ANTIA OH MEOH OH 2 PEAKLIST 030407_ANTIA 2 Tissue Mass Volume 51 6 38 2 3 Genotype ID At Ler Cell Suspension At Ler Cell Suspension At Ler 4 Organ Cell suspension Cell suspension Ce Treatment ID None None 6 Treatment Duration 0 0 7 Treatment Dosage 0 0 8 Replicate 1 2 ers Analyte Signal ID Retention m z Time min es Kovats 10 L Alanine 2TMS ID14 RI1079 6 MZ117 12 639 1080 8 117 53734 52579 58070 90035 9 11 L Alanine 2TMS ID15 RI1081 4 MZ117 12 641 1080 9 117 53734 52579 58070 90035 9 12 Glycine 2TMS ID18 RI1106 4 MZ204 13 248 1106 8 204 716 898391
33. quantifier ion specified ead ae 3 1045 1 Lactic acid 2TMS Lactic acid Matched this library a entry will not Foto be matched Done a As shown in the screenshot above the Analyte Name column shows the name of the library entry The Metabolite Name column shows the metabolite name that the string processing algorithm has derived from the Analyte Name after removing from it all the strings recognised as derivative information The Metabolite Name Matched column indicates whether the derived metabolite name was found in the MetabolomeExpress Metabolite Name InChl adapter database To ensure that library entries for unknown metabolites are never identified as known metabolites it s a good idea to enclose their names in square brackets as done by the clever people at MPIMP in their library The Library Entry ID column shows the IDs of the entries which in the case of this freshly imported library have not been set The Quantifier lon s column shows the quantifier ions specified for each entry Again these need to be carefully selected and entered using a spreadsheet The final column provides buttons that load the mass spectrum of each library entry into either the top or bottom MS display window You may need to expand the MS Comparison window if you have already collapsed it This is useful for examining and comparing library spectra Below you can see a comparison between the spectra of Lactic acid 2T MS and Alanine 2TMS
34. reng nica Ha rd pd a oer retentis teen it each DCS IO ghe 75 Seer Viri rg lschig tr Cent coin deed Est elor velin fala oe cee elder boned Los reg Cent ee rat ee eee CH pir f a Breu eee pd rege DE tiama TR PET Thin is jest os tha mat MATRE gt asap hat thn pan arena In and lu gien gh of the mathe have bean narmalud t the manimems peek moun be that signal aevo ul tha dein lias in ha oak Tharelarn d valuta aoo Gens pe Bach Est 15 Spipesien beehives 1d aul 1 Za barraat ii arimba her cosas virile ir Girl eerie ani rosa daa seht Reni ops thee via iod iiem dehineg min BR tee HA TRIS Thin in pork Hle foa ATMA TRL format erst irata c hare a paek area ini aac of tha cons oda cl the matrix t har doa paak pen fa rg er en tree and tha renton inda Gegen De chperved axpecisd ET mpare o Sagres DI by Reread aahi UL Thal eui Farina cues ot currently Ea uiid far Foster occides unag Lge beatin omi fer Dngncdng pouitis peciblacad raised bo pak datacfon auch M g eent Al minga in MER Seene EN eg 3 AEct 2Ldaarogbc TELE TAN Thai a Pha Fora inqsorbed Froe fog pkr baiia pack ace Molecular Prolier Tt ieoch licae ha met borrat bt een bry deeds Porton headers other that daa Tia r the top of each coluen The format a DCNS D Dith eot 15 gt Hiii mci wohl Let gung scar lur gered deefe te Alerts gata dete aca G ke arra HS eg Ta Normalisation Overview Your mate amp caled GCM5 ID Batch Report_19 5e er 2008_B 8 21 normareaMATREX You can download t u
35. validated This user s manual was compiled by Adam J Carroll Copyright 2010
36. 0 n docosane 3 42 454 2800 100 100 n octacosane 6 46 802 3200 100 100 n dotriacontane H 51 014 3600 100 100 n hexatriacontane 8 M a gt M Alkane RI 030407 Oh J 4 Ready ESO D Once you have created the CAL file upload it into the experiment folder by FTP Assuming you have uploaded an appropriate MSHI library to one of your library folders or have used a standard GC MS protocol that allows you to use one of the public libraries provided by MetabolomeExpress or one of its users you will now be ready to perform MSHI library matching as soon as your data data import and peak detection is complete NOTE If your GC MS data was acquired over more than one batch sequence for example if 10 runs were done one week and 10 runs done the following week there may be significant systematic differences in retention time between the different batch sets Therefore you will need to create one CAL file per batch sequence to ensure calibration is always correct If you want to make direct comparisons between different samples it is important to run the samples to be compared in the same batch ie don t run your treatment samples one week and your control samples the next week for instance 1 4 2 Choosing a research area and importing sample information Once you have uploaded basic sample information in the form of a completed MINIMET TXT file you can use the file to generate a functional template METADATA TXT file appropriate to yo
37. 2 616 3324851 E 13 Glycine 2TMS ID18 RI1106 4 MZ147 13 248 1106 8 147 3574 656332 2788 918602 1 14 Glycine 2TMS ID18 RI1106 4 MZ102 13 244 1106 8 102 6371 890846 4909 9702 2 15 n dodecane ID29 RI1200 MZ41 15 412 1200 41 19175 21 25237 31916 3 16 n dodecane ID29 RI1200 MZ57 15 413 1200 57 43435 01169 56827 49116 7 17 n dodecane_ID29_RI1200_MZ71 15 413 1200 71 26995 63503 35311 65513 4 18 L Valine 2TMS ID35 RI1211 4 MZ144 15 689 1211 9 144 12008 70837 12205 30443 2 19 Unknown RI1231 2 ID38 RI1231 2 MZ234 16 175 1232 234 216 1111958 173 5470149 1 20 Benzoic acid 1TMS ID41 RI1254 3 MZ179 16 664 1253 8 1245 599949 1928 191292 2 IR gt GCMS ID Batch Report 10 Septemb J 2 Re analysis of publicly disseminated data using Database Explorer The Database Explorer module provides tools to interact with data in the MetabolomeExpress database of metabolite response statistics It currently contains four sub modules Database Statistics ResponseFinder MetaAnalyser and MSHI Library Manager The latter has been described above in the section Building and using custom MSHI libraries The other three are described below 2 1 Getting an overview of the database with the Database Statistics panel The Database Statistics panel provides a summary of the current contents of the MetabolomeExpress database of metabolite response statistics It currently displays two tables one summarising the total amount of data in the database
38. 8 d 1 i L H i 3 E 2 4 Identifying phenocopies using PhenoMeter in development Fiti ss Most biologists are familiar with BLAST search algorithms which allow you to submit a DNA RNA or protein sequence as bait and retrieve sets of homologous sequences scored and ranked by similarity from a large database The PhenoMeter is an analogous tool that lets you use a metabolite response ie a set of metabolite fold changes and p values for a particular class comparison as bait and retrieve sets of other responses from the MetabolomeExpress database that are ranked and scored according to their similarity The interface for the PhenoMeter tool is currently exactly the same as the MetaAnalyser Metabolite responses in the database are represented in a tree structure which begins at the publication level and branches down into the experiment level and then the class comparison level You can make selections at any level using the check boxes provided However you are advised only to select one or two class comparisons per search in order to avoid excessively long query times The PhenoMeter control panel is shown below PhenoMeter Control Panel Please select metabolite responses from the tree D Giraud E Ha LH Clifton R Carroll A Estavillo G Tan YF Howell KA lvanova A Pogson BJ Millar AH Whelan J 2 wilson FB Estavillo GM Field KJ Pornsiriwong VW Carroll AJ Howell KA Woo NS Lake JA Smith SM Harvey Mil
39. Description A valid species name Must be at least 2 words with no numbers Species name will be checked against the NCBI Taxonomy database and a warning issued if the name is not recognised Valid species names may be restricted to a particular branch of the taxonomic tree using an optional configuration parameter explained below Configuration Parameters Valid species names may be restricted to a particular branch of the NCBI taxonomic tree using an optional configuration parameter NCBI taxon rank value eg superkingdom bacteria Code Examples Example1 v sp Enter any valid species name Example2 v sp superkingdom bacteria Enter the full name of any bacterial species Example3 vj splgenus saccharomyces Enter the full name of any oaccharomyces sp Example Valid Values Example 1 Eucalyptus globulus Example 2 Escherichia coli Example 3 Saccharomyces cerevisiae Example Invalid Values Example 1 E globulus Example 2 Homo sapiens Example 3 Amanita phalloides Controlled Vocabulary cv Data Type Code cv Validation Description A single term that must match to a term specified in the configuration parameters Configuration Parameters A forward slash separated list of allowed terms A number of controlled vocabularies are built in to the MetabolomeExpress database and these may be included in the allowable term list by including as one of the allowable terms their code name enclosed in square brackets see
40. Disease Ontology http do wiki nubic northwestern edu index php Main Page terms for human human pathology diseases Obtained from http obo cvs sourceforge net checkout obo obo ontolo henotype Pathology human disease obo Mouse Pathology Ontology http eulep pdn cam ac uk Pathology Ontology index php terms for mouse_pathology mouse pathologies Obtained from http obo cvs sourceforge net checkout obo obo ontolo henotype mouse pathology mouse pathology obo E jf and WAY Units of Measurement Gram based units of mass without abbreviation eg grams NOT g Unabbreviated gram based units of mass with a specification of either dry units tissuemass weight or fresh weight eg milligrams fresh weight NOT mg FW Internal Reference ir Data Type Code ir Validation Description Field value must match with at least one value given for another metadata field specified by the configuration parameter This validation method is used to ensure consistent naming of elements in the metadata file For example the Extraction Protocol ID given for an extract in the EXTRACTS section should match to at least one of the values for Extraction Protocol ID given inthe BIOSAMPLE EXTRACTION PROTOCOLS section Configuration Parameter The name of other metadata field to search for a matching value Given in the form METADATA SECTION Field Name Example Code v ir BIOSAMPLE EXTRACTION PROTOCOLS Extract
41. E F G H Analyte Name Metabolite Name InChi Identifier Chemical Class is_unknown_structure is_quant_peak is_artefact is_internal_std InChl 1 C9H804 c10 7 3 1 6 5 8 7 11 2 4 9 12 13 h1 5 10 11H H 12 13 f h12H 3 4 Dihydroxycinnamic acid 3TMS 3 4 dihydroxycinnamic acid Aromatic Acids 0 1 0 0 InChi21 C4H9NO2 c5 3 4 Aminobutyric acid 3TMS Gamma Aminobutyric acid 1 2 4 6 7 h1 Amino Acids 0 1 0 0 3 3 5H2 H 6 7 InChi 1 C6H13010P c7 2 1 16 17 13 14 15 3 8 4 9 5 1 6 Phosphogluconic acid 7TMS 6 Phosphogluconic acid 0 6 11 12 h2 5 7 Sugar Phosphates 0 1 0 0 10H 1H2 H 11 12 H2 1 3 14 15 t2 3 4 5 m1 s1 InChi 1 C5H5N5 c6 4 3 Adenine 2TMS Adenine 5 9 1 7 3 10 2 8 4 h1 Purines and Purine Derivatives 0 1 0 0 2H H3 6 7 8 9 10 InChi21 C22H46 c1 3 5 7 9 11 13 15 17 19 21 n docosane n docosane 22 20 18 16 14 12 10 8 Alkanes 0 1 0 1 6 4 2 h3 22H2 1 2H3 InChi 1 C5H1205 c6 1 Ribitol STMS Ribitol 3 8 5 10 4 9 2 7 h3 Alcohols and Polyols 0 1 0 1 10H 1 2H2 t3 4 5 Pentasiloxane dodecamethyl MS92 RI1152 7 9 Unknown Probable Disaccharide RI2748 9 Unknown Unknown Unknown 1 1 0 0 1n Unknown Unknown Unknown 1 1 1 0 Once you have appropriately edited your Analyte Annotations Table save it as a tab delimited text file and place it either in your libraries FTP folder or in the folder of the experiment you wish to apply the annotations to If you put it in the libraries f
42. Heport to Data Matrix tool includes options 6 1 2 Raw data assisted missing value replacement When a data matrix is constructed there are almost invariably cases where a particular metabolite has been detected in some samples but not in others This gives rise to missing values As many popular multivariate analysis techniques such as PCA and HCA cannot deal with missing values it is necessary to fill these with some kind of proxy value Many tools that carry out data matrix construction will either leave the missing values blank or will try to mathematically impute the real value Alternatively some will set all the missing values in a matrix to some low number based on the assumption that the value was missing because the compound s signal was below the baseline Sometimes this is a valid assumption but with most mass spectral library matching based approaches it is not Quite often missing values arise because perfectly valid and clear signals did not for some reason quite get past a stringent library matching filter Therefore an increasingly popular alternative to all the previously mentioned approaches is to use chromatographic information from cases where the library matching led to positive matches to locate the missed signals in the raw data for which library matching failed This is the approach used by the MetabolomeExpress Match Report to Data Matrix tool When this tool encounters a missing value it determines from the sign
43. IST 030407 METHANOL OH 3 PEAKLIST 030407 METHANOL 0H 4 PEAKLIST 030407 METHANOL 0H 5 PEAKLIST Submit Sample Selection Make your selection and click the Submit Sample Selection button A progress bar will then appear in the Output window while the matrix is assembled This process can take a little while if there are lots of files NOTE If you want your peak areas to be normalised to sample mass volume and properly annotated with sample class information you will need to provide this information by having a functional METADATA TXT file in the experiment folder The easiest way to create one of these is to import a basic set of sample information from a MINIMET TXT file see Step 4 in Section 1 4 Uploading and managing your datasets via FTP If no metadata is provided samples will be assumed to have equal mass volumes and class attributes in the matrix column headers will be set to Unknown as in the screenshot below me Navigation Database Navigatien aC Panin s cgPantEn tapoiliome angyiliiagy_inigenal Grm ngogs Public 34 G Arabidopels ACTA KO Moderate High Lig 3 Arabidopsis Call Suspension Timecourie aj CH Arabidopsis GST RMA SA and HzOr2 Titar a Arabidapkis Milochenaial GTC Carmglax i H Al Arabidopsis HMipcheno ETC Compan H a4 Arabidopels SAL 1 Mutants abit and rt zi 3 Emi Experenen arasidopsis Cells An 37 030407 ANTIA H MEOH OH 1 6OF Si Em 030207 ANTIA H MEOH
44. IST files will be generated Check this box if you want to overwrite old XIC files and old PEAKLIST files Select data file s for import and or peak detection here Any CDF files selected will be imported and used to generate PEAKLIST files if Extract Peaks is selected Any XIC files selected will be used to generate PEAKLIST files if Extract Peaks is selected Otherwise they will be ignored 5 MSRI library matching 5 1 How to conduct an MSRI library matching process MSHI library matching is defined as the identification of mass spectral signals corresponding to target analytes in our case metabolite derivatives in a GC MS data set by matching detected signals to entries in library of mass spectral and retention index information for those target analytes an MSHI library If you click on the tab entitled MSAI Library Matching you will be presented with a screen something like this me etabolome X D TG E amp TH de D N i D xHf FEIFECOR ASH i a Messen Vi Rum Data Yese Duca Be cvet aed Peak Deezee Mani St Making Sues aed Den Epir als tput Thus mt wbi peu Warte maiia Saad by mahr paik bei bant Verr iri of iiiar Kr i euer eee ebore Library Hacckesg Contre Panel e neben yras D fae DIDIT ANTIA DH MEOH DH T PERELIST CODAUT AMTIA DH MEOH 0H 2 PEARLIST C007 AMTIA OH MEOH 0H 3 PEARLIST OZ AMTIA OH MEOH OH 4FEAXLIST Dat AMTIA M MECH OMS PEAKLIST 6 Macching Criraria Ri winden d Ww MST merged diuaecu 10 Let RI Unm Pi
45. Inert MSD quadrupole detector Derivatization products were injected in the splitless mode and separated over a 30 m Varian FactorFour VF Sms column with 0 25 um thick 95 dimethyl 5 diphenyl polysiloxane film and 10m integrated guard column Varian Inc Palo Alto CA USA Product No CP9013 using a linear temperature gradient before positive mode electron impact ionization at 70eV and detection using a single quadrupole mass selective detector scanning from m z 40 to 600 Spectra were extracted from raw data files by deconvolution with freely available AMDIS software Kovat s retention indices RIs were calculated in AMDIS using external calibration with a series of n alkanes run alongside samples used for library construction For details of the derivatisation procedure see ANALYTICAL SAMPLE PREPARATION PARAMETERS For details of the instrument method see INSTRUMENTAL PARAMETERS For details of spectral extraction by software data processing see DATA PROCESSING PARAMETERS Library spectra obtained from biological samples and not verified by authentic standards are indicated by enclosure of their Analyte Name in square brackets The Analyte Names of these entries are generally of the form Name of best match to the NISTOS MS library MS AMDISMatchScore Rl Kovat s Retention Index In some cases where the NISTOS match is suspected to be correct names include the comment Putative but Unconfirmed These should be interpreted with caution
46. LIS D nen ANTIA DH MECH DH 3 XI B30407 AMTIA GH MEOH OH A CO 30407 ANTIA OH MEGH DH A PEAKLIS 030407 AHTIA GH MEOH DH A ai 030507 AMTIA OH MECH OH 5 CDF D30407 AHTIA QH MEOH QH 5 PEAKLISC 5 3 s07 AHTIA OH MECH DH 5c z ES ES Ez E E id 030407 METHANOL OH 1 CDF 030407 METHAHOL 0H 1 PEAKLIST 030407 METHANOL OH 1 00 030407 METHANOL DH 2 Cor 030407 METHANOL OH 2 PEAKLIET 030407 METHANOL OH 7 XIC 630807 METHANOL 0H A COF 030407 METHANOL OH 3 PEAELIST 030407 METHANOL 0H 3 XIC 5 030607 METHANOL oH 4 COF Risen m Contact Information Lkenseg Fint nbitel 130407 AHTIA GH HE DH OH T PIEAELIS 304807 ANTIA GE MEOH OM Z PEAKLIS E ben t Previous teg A Experiment Explorer Arab deqes GST RAMAI SA ard H202 Treatments m Database Expleeer ep Help p Recetten Statistics and Data Eplorakicn Output Statistics and Data Exploration Control Panel Jeer authoraben OK d Home Mataleta Viewer Raw Data Viewer Data Import and Paak Detection MERI Library Matching DATA MATRICES The folowing data matrices have been generated wing the sample nfermatian that you entered on the prewous screen These may be downloaded using the inks provided These matrices were buit in a total of 77 seconds Total number of missing peaks filed by raw data assisted mesng value replacement A80 peaks Farm il Diomnipad Link HALTEZ EE Lossen huen ka rain
47. Min iin Hp i wit Ha Ne Ha Ha Mo Ha No Treatment Treatment Treatment Treatment Treatment Treatment Treatment Treatment Treatment Treatment Tri Ha Ha Ho Ha Ne No Ha Ha Ha Les Treatreent Troatrment Treatment Treatment Treatment Treatment Teeabrment Treatment Tre z 3 5 6 T a 10 Si e HE SEH E d SET s 1 3 ET T ETET L DEN See p aminaethan RSS 3 Rod M7102 jor ow om Do 32 EE ine 3 trimethylsilaxy MSR De 4 ID3 RIT000 4 Mz152 107241000 15 BELT EB IC TELLURE aart Cas E i Lactic acid met wes wire A P utative but Unconfirmed IDG RITUZS 1 M TAK 11 402 1078 1 147 0 03 NR LE A 1 34 D Alanine H trifluoroacetyl trimethylsilyl ester MS85 RHO44 6 ID7 RITO446 MZ242 11 825 1045 242 EE 23 15 P L Alanine 7TMS Ibid Rid z9 6 Mii 12 639 1080 8 117 ANE E i1 41 L Alanine TMS ID RIIORT A HI 12 541 1080 9 117 RE S id 11 Phosphonic acid methyl bisptrinmethylsilyl ester M552 RI1031 2 Hp RIO MZX32 12 833 103 4 332 02 iS rs 41 Glycine 27M S ID18 RIT106 4 MEM 131 2481106 2 204 O3 E 011i 04 018 025 OH 055 AE Glycine 2 TM3 ID18 RITIOG4 MZTAT 13 2401106 8 147 EDEN 013 Odz 025 D 0 2 059 52 Giycine ER 1018 RITIOS A4 Mz102 13 244 1106 8 102 MENO NN oi TG RER 32 021 DS 045 ER N Trimethylsity 2 pyrrolidinone MS88 R11137 1 ID20 RIT137 1 MZ142 EAR um 035 Esc oa we om uz as SS oes Pentasila
48. Panel Match Beport bo Gala Matris Daba Mabir Eruca ors Biasit Explorer Coos aber S Ea c ben BZA HEA Coreiabeen Arlee ecc ei icher Beeseeebseier My i DB Shake laeri Please sckct a MATCHRIPORT file GEMS ID Batch Repor_11 August2008_5 1447 MATCHREPORT e Mermsise to Inter Seandard Thiberia So ne f Tf bi fer Sirig reel Fe anie part of See mierna nn e makeh rna Fibstal Upload Hefe Contact bahar Fulton Lane The Statistics and Data Exploration Control Panel is where all statistical analysis procedures are initiated The results are displayed in the Output window 6 1 Construction of data matrices from MSRI library matching reports Once MSRI library matching has been carried out the next step is to assemble a data matrix from the MSRI library matching report This process arranges all the results in the match report into a table where instrument runs are represented as columns and the various detected signals are represented as rows You will require write permission on the repository containing the dataset to do this If you are analysing someone else s public dataset for which you don t have write permission they will most likely have already created a data matrix for you 6 1 1 Some notes on normalisation and quality control To control for pipetting errors and variations in starting sample mass volume data matrix construction usually involves normalisation to some internal standard peak area and also to tissue mass vo
49. TMS ID14 RI1079 5 MZ116 1284 A 1 26 E 2 59 L Serine 37MS _IO65_RI1359_MZ218 1 0 17 19959 179 4 29 1 47 Amino Acid L Valine 2TMS ID35 RI1211 4 MZ144 15689 12119 12 ALE x 824 12 12769 1 24 i 1 1 1 Amino Acid L Tryptoph n 3TMS 1D262 R12210 3 M2291 1 Aminocyctopropanecarbonftic acid GIS D33 RI1205 4 MZ202 Amino Acid 1 Aminocydopropanecarboryic acid a me ID32 RI12047 MZ202 Glycine N acetyt trimetnitsihi ester MS71 RI1227 511037 R11227 5 M2174 L Alanine N trifuoroacetyt mmetnytsily ester Amino AGG MS85 RI1044 6 107 RI1044 6 MZ242 You can download statistical reports in various formats using the hyperlinks at the top of the page If you see an interesting statistical result try double clicking on the analyte abundance ratio to see the underlying GC MS signals 6 5 Principal Components Analysis PCA The Metabolome Express PCA tool provides you with the ability to very quickly find interesting patterns in datasets on the server The PCA tool utilises sample class information contained in the experimental metadata file to provide publication quality 2D and 3D PCA score plots together with sortable variable loadings heatmaps and scree plots All you have to do is select a matrix specify whether you want to include redundant signals repeat measurements of the same metabolite but using different GC MS signals two different quantifier ions for example and whether to include unknown metabolites that
50. The MetabolomeExpress User s Guide v 1 0 By Dr Adam J Carroll QUICK START Load a dataset by left clicking or right clicking or control clicking if you are using a single button mouse on an Apple computer on the experiment folder and left clicking Load Folder Contents Navigation d Database Navigation e j CH PlantEnergyBiology_ Internal 3c PlantEnergyBiology Public d C Bb sranidopsis AOX1a KO Moderate High Light and Drought Stress 3 C A Arabidopsis Cells Rotenane Timecourse d C i Arabidopsis GST RNAI SA and H202 Treatments d C A Arabidopsis SAL17 Mutants alx8 and fry1 1 d bal r EEEE mtETC Cplx I Mutant ndusf4 Cmplmnt day and night H CJ Arabidopsis mtETC Cplx Mutants ndusf4 and ndufat night 3L a Ie rxnerimenrt m Load M Contents import sample Information From MIMIMET TXT P zd C 4 Rice Germination Timecourse Air vs N2 d Lal A Rice Germination Timecourse Air M2 Gas Switch E ia Rice Germination Timecourse M2 Air Gas Switch S Table of Contents 1 MetabolomeExpress Getting starte 4 UN MER 0e cc Ro I II N eee eee ee eee te eee te oe er eee ee eee 4 1 2 Public vs private lee 5 1 3 Obtaining your own MetabolomeExpress data repository Registration 5 1 4 Uploading and managing your datasets viakIR 6 BC Ze MN erosut mt 6 step 1 Create a data folder for the experiment seessseesse 6 Step 2 Upload Taw GO MS files E 6
51. Trimethylsiloxy 2 trimethylsilylaminoe I Trimethylsiloxy 2 trimethylsilylaminoe Pyridine 3 trimethylsiloxy MS80 RI10 Pyridine 3 trimethylsiloxy MS80 RI10 Pyridine 3 trimethylsiloxy MS80 RI10 Pyridine 3 trimethylsiloxy MS80 RI10 Pyridine 3 trimethylsiloxy MS80 RI10 Pyridine 3 trimethylsiloxy MS80 RI10 Here s what the example network looks like when imported into Cytoscape ver 2 6 0 and laid out using the Organic layout Fie Edt View Select Layout 8 aagea Poges HAR een ggg E NEM o Os FE GCHS ID Batch Report_11 August 2008_5 14 47_CYTOSCAPE_CORNET_19 August 2008_ 4 49 35 TXT ip Network vieManper Cerebral E j Q0 ECKER ay G i EECC oS eoe e e S226 k E Node Attribute Browser Edge Attribute Browser Natwork Attribute Browser Parallel Coordnates Welcome te Cytoscape 205 0 Bitci drag t ZOOM Middle thek dra fo PAN Here s a closeup Adenosi S phy Je e se Notice the classic robust correlation between Fructose 6 P and Glucose 6 P in the top right corner 6 8 Submitting a dataset to the main database of metabolite response statistics Once you are happy with your data processing and have possibly even published your results you are strongly encouraged to submit your dataset for indexing in the main database
52. XIDASE 1a in fiere mre Results 2008 04 18 NE ES 6 Mitochondrial Alternative ne GAL cok d Leaf Mod ewech SEN Light 166 in Acute Sensitivity to Combined Light and WS Oxidase 1a KO Arabidopsis Naana e ALK rougnt F ign Lig Drought Stress Plant Ae Vol 147 pp 595 6 Wilson PB Estavillo GM em KJ Pornsiriwong W Carroll AJ Howell KA Woo NS Lake JA Perreira F Smith SM Harvey Millar A von Caemmerer 5 aracterisation Pogson BJ 2009 The 2008 12 16 desi dem s Mre tolerant alx8 and emie C24 M 11 Leaf NormalGrowth 188 nudeotidase phosphatase SAL 1 is a negative n fry1 1 Arabidopsis mutants ana regulator of drought tolerance in Arabidopsis Plant Journal 58 2 299 317 Garmier M Carroll AJ Delannoy E Vallet C Day beier ebay Timecource Treatment of 2 2 Finding experiments of interest using ResponseFinder The ResponseFinder module allows you to search the MetabolomeExpress database for metabolite responses of interest based on metabolite name minimum fold change maximum p value metabolite response directionality species and organ The screenshot below shows the ResponseFinder control panel set up to find any results where 2 Oxoglutarate was observed to be increased or decreased by at least 2 fold and a p value of 0 05 or less in any organ of Arabidopsis thaliana Kel Home l A Experiment Explorer lig Database Explorer e Help amp Registration Database St Statistics I ResponseFinder lL MetaAnalyser
53. ails D 970 3 10 118 ass vere No Ri Hits 870 8 10 125 mun ru No Fa Hits Drog 10 111 aac Ho Ri Hits 9 1 013 sceau Grund Ha Ri Hits 9714 10144 ien Doer No Ri Hits gc 10 15 immer e iru No Fo Hits ora 5 1B 17 vn ean Ess Mo F3 Hil 9727 10 178 Arte iin tee T Ne Ri Hits 3 n ren rotes Dagiay EN ga gem aba Fun Es el Pe 1I Lim UT EEFT ESOUR Pt UE ee PESSLD 8 TA APqPSUERECCUSE GU ee n UACQCUTESDOA SS EUN OW TSORISCTWODTOWSOP FW tA Lia Pasa Ger ART der L fes UA d mpm ATTE Boe ts 7 B ED chr mr gr d Ferma s AT dO Lavery Hasrhesg Cone Panel Ds Dik kg RT Lt Sec giae WT dr vd Ric hini TAS EF kt KN aT ie ler tz me ee UE dea wc Han Spectral Coampariaces You can probably tell by looking at the table that each row corresponds to another 0 1 RI unit bin of EIC peaks an MST by our definition None of the MSTs you can see in the screenshot above had an RI match in the searched RI library they are all weak intensity MSTs made up of a few tiny EIC peaks However if you scroll down you will begin to see MSTs with something in the Hl He column If the name of the RI Hit is black it indicates that the MST RI falls within the RI Tolerance Window default 2 RI Units of the corresponding library entry but the identification wasn t supported by either the presence of a library specified quantifier ion or by similarity between the ion ratios of the MST and ion ratios of the library spectrum If th
54. al annotation which m z channel was used as the quantifier ion in that row of the matrix and uses the integration start times and end times from positive matches in the library match report to determine the average peak start and peak end retention times of the signal using only retention times from runs acquired in the same instrument batch sequence according to the Batch Sequence ID values for the Analytical Runs in the _METADATA TXT file It then reads the raw data file showing the missing value and integrates the appropriate m z signal from the average start time to the average end time and places this value in place of the missing value Values obtained in this way are flagged with an asterisk in the resulting tab delimited mzrtMATRIX file You can always check the peaks later with the raw data viewer to convince yourself that the numbers make sense based on manual interpretation of the raw data IMPORTANT NOTE When an mzrtMATRIX file is submitted for comparative statistical analysis class comparisons for which at least one class had more than 5096 missing values will be excluded from the result set and replaced with an X 6 1 3 How to build a data matrix using the Match Report to Data Matrix tool Matrix construction is initiated using the Match Report to Data Matrix tab Choose your input MATCHREPORT file here Statistics and Data Exploration Control Panel Match Report to Data Matrix Data Matrix Renormaisation Matri
55. and display of data for statistical analysis Instructions for their use are given below 1 5 5 Using Analyte Annotations Tables to customise data filtering If using custom MSHI libraries most of the tools in the Statistics and Data Exploration module of Experiment Explorer require that an analyte annotations table file containing annotation information for the entries in those MSHI libraries be present either in the libraries FTP folder of the repository containing the dataset of interest or in the actual folder of the individual experiment To generate a template analyte annotations table file from an MSRI library Display and Validate that MSRI library in the MSHI Library Manager as described above You may then scroll down and use the hyperlink to download the automatically generated analyte annotations file already containing the following automatically assigned annotations for each library entry Metabolite Name The common name of the underivatised metabolite matched to the Analyte Name of the MSRI Library entry most of these common names are the same as those in the Human Metabolome Database InChl Identifier This is the unambiguous structural identifier string of the underivatised metabolite corresponding to the library entry if it is known Chemical Class This is the chemical class of the underivatised metabolite eg Amino Acids most of these classes are as defined in the HMDBJ is unknown structure his is a boolean w
56. anel Rice Germination Timecourse Air vs N2 Please select one or more of the following data Peak Detection Settings 3j C ndusf mtETC Complex Mutant Complem files i doe ruht 5o e lt 030407 ANTIAOH MEOH OH 1 XIC 3 CJ ANU RSB MS Fadility Public 030407 ANTIAOH MEOH OH 2 XIC Min Peak Area 1000 030407_ANTIA OH MEOH OH 3 XIC Min Peak Height 500 l 030407 ANTIA 0H MEOH 0H 4 XIC Ee Se SE em 030407 ANTIA OH MEOH 0H 5 XIC M TON det 2 030407_METHANOL OH 1 XIC mr m 030407 METHANOL DH 2 XIC es Widih 15 030407_METHANOL DH 3 XIC TM 030407 METHANOL 0H 4 XIC Extract Peaks O v 030407_METHANOL OH 5 XIC M Over Write d RS B leor Refresh j Contact Information Licensing Source Code SSC IIE 4 1 A Guide to the MetabolomeExpress PeakFinder Algorithm The MetabolomeExpress PeakFinder algorithm is responsible for the detection and measurement of chromatographic peaks in extracted ion chromatograms EICs When a raw data file is sent for peak detection the PeakFinder algorithm is passed each nominal mass integer mass EIC in the data file as two equal length vectors signal intensity and retention time one by one until peaks have been detected in all EICs The end result is a tab delimited PEAKLIST table with columns for m z retention time peak area peak height peak width integration start time and end time the scan numbers of the integration start and end points the intensity of the signal at the integ
57. ation required for signal normalisation and statistical grouping This information is provided in the form of a simple tab delimited table which must be called MINIMET TXT which stands for MINImal METadata This simple metadata file can be used to construct a starting template of the somewhat more complex METADATA TXT metadata format This allows you to process your data and get statistical results quickly without having to spend time completing the larger _METADATA TXT file Many of the fields in the larger format are not required for data processing but are required for proper Metabolomics Standards Initiative MSI compliant public dataset dissemination The MINIMET TXT format has 6 columns which must be labeled Sample ID Genotype Treatment Organ or Biomaterial Type Timepoint and Sample Mass or Volume The order of the columns is not important but the column headings must be exactly as shown Each row represents a single GC MS run The information to put in each column is as follows Sample ID This is the name of the NetCDF GC MS data file without the CDF extension Genotype This is the short name of the genotype of the organism that was analysed in the sample Make sure that all samples of the same genotype have exactly the same entry here Try to use a well established short name if there is one Treatment his is a short descriptive ID for the experimental treatment applied to the organism analysed in the
58. atrix Data Matrix Renormalisation Matrix Explorer Comparative Statistics PCA HCA Correlation Analysis ResponseFinder ReponseFinder MySQL DB Stats Import Please select a data matrix for Correlation Analysis GCMS_ID Batch Report_11 August 2008_5 14 47 mztMATRIX_RENORM v Enter the threshold correlation coefficient for node connection 0 7 Filter redundant analytes Remove analytes of unknown chemical structure Upload Wait a few moments while the correlations are calculated and the network graphs are constructed You will then be presented with a page like this Correlation Analysis Progress Reading Dataset GCMS ID Batch Report 11 August 2008 5 14 47 mzrtMATRIX RENORM Output CORRELATION NETWORKS This format may be imported directly into the network visualisation and analysis software Cytoscape It contains all analyte analyte correlations with absolute correlation coefficients above the threshold that was specified in the submitted request It includes two attribute columns giving the absolute correlation coefficient and the sign of the correlation ie Positive or Negative This allows the user to take advantage of Cytoscape s filtering and attribute to visual style mapping features This format may be imported directly into the open source network visualisation and analysis package Pajek It contains all analyte analyte correlations above the specified threshold Cytoscape Pajek Correlation Matrix DNA DEAE De
59. aw Data Viewer Raw Data Import and Peak Detection MSRI Library Matching Statistics and Data Exploration P Database Explorer Database Statistics ResponseFinder MetaAnalyser MSRI Library Manager wy Help Documentation 2 Registration Registration Application Form Figure 1 Structural overview of MetabolomeExpress 1 2 Public vs private data access C MySQL DATABASE Statistical Results Table Metabolite response statistics Imported from FTP repository upon request Dataset must be complete Metadata must pass validation Metabolite Name InChl Adapter Table Links synonymous metabolite names to their common InChl identifier References over 100 000 metabolite names Curated by MetabolomeExpress InChl Metabolite Information Table Provides biological and chemical information on 7900 metabolites Indexed by InChl identifier Built from the Human Metabolome Database Curated by MetabolomeExpress Ontology Tables Anatomy e Fungal e Plant Development e Human Disease e Mouse Arabidopsis thaliana Escherichia coli Drosophila melanogaster Homo sapiens Oryza sativa Saccharomyces cerevisiae NCBI Taxonomy MetabolomeExpress houses both public and private data including GC MS libraries Public data is accessible to anonymous users ie users automatically logged in as guest via the web interface Private data may only be accessed by registered users who are logged in and have permission to access the private data s
60. ble click on an analyte signal name the left most column you can be taken straight to a chromatographic overlay of that analyte s quantifier peak in the selected files Whether a column header is set to either blue or red in the matrix view will determine whether the corresponding GC MS signal is plotted in either blue or red in the Raw Data Viewer Columns with white headers will not be plotted Below is an example Select the headers so they look like this here we have set the 3 hour Antimycin A treated samples to red and the 3 hour methanol treated control samples to blue 48 1 50 5 38 8 34 8 34 8 55 54 7 49 42 6 At Ler Cell At Ler Cell At Ler Cell At Ler Cell At Ler Cell At Ler Cell At Ler Cell At Ler Cell At Ler Cell i Suspension Suspension Suspension Suspension Suspension Suspension Suspension Suspension Suspension Cell Cell Cell Cell Cell Cell Cell Cell Cell 1 suspension suspension suspension suspension suspension suspension suspension suspension suspension AntimycinA 3hAntimycinA 3hAntimycinA 3hAntimycinA 3hAntimycinA 3hMethanol 3hMethanol 3hMethanol 3hMethanol 3h 3h 3h 3h 3h 3h 3h 3h 3h 3h 1 1 1 1 1 1 1 1 1 2 3 4 5 1 2 3 4 0 09 L Ion 059 065 0o72 o1 0 f 029 oa 012 oo on oo 028 on or or on or oo on f ow osm Cee Lee ee Lis on ow os o oa o o2 so SE Now scroll back to the analyte signal name column and double click one of the signal names
61. breaks threshold in minutes Integration End Time The retention time at which the end of the peak is reached in minutes Total Peak Area The total area under the peak from start to finish in arbitrary peak area units Peak Height The height of the signal at the peak apex in arbitrary peak area units Peak Start Intensity The height of the signal at the peak start retention time in arbitrary peak area units Peak End Intensity The height of the signal at the peak end retention time in arbitrary peak area units Peak Purity Factor The ratio of the total peak area to the area lying under the lowest integration point Peak Base Area The total area lying under the lowest integration point Number of Scans The total number of scans between the peak start retention time and the peak end retention time Peak Start Scan Number The scan number at the peak start retention time Peak End Scan Number The scan number at the peak end retention time If you include these headers exactly as written at the top of the columns in your peaklist file MetabolomeExpress will recognise them and you may have the columns in any order If you omit the headers MetabolomeExpress will assume you have used the default column ordering ie the list order of the columns listed above The MetabolomeExpress PeakFinder algorithm automatically outputs peak lists with these headers For a peak list to be linked to a chromatogram it MUST be named w
62. chical Cluster Analysis HCA The HCA tool currently provides two modes of clustering 1 Bi clustering of the Metabolite x Sample Data Matrix and 2 Bi clustering of the Metabolite x Metabolite Correlation Matrix Both modes of clustering are performed using automatically generated R scripts that utilise the heatmap function freely available as part of the Bioconductor open source bioinformatics software package for R see www bioconductor org The output is generated as a PDF which is displayed directly in the Output window You will therefore need to ensure you have installed a PDF viewer browser plugin if you wish to view the clustering results directly in the browser Alternatively you may wish to download the PDF document using the hyperlink at the top of the Output window and view the PDF using your normal PDF viewer To see a nice example set the HCA control panel as shown below and hit Upload Statistics and Data Exploration Control Panel y Match Report to Data Matrix Data Matrix Renormalisation Matrix Explorer Comparative Statistics PCA HCA Correlation Analysis ResponseFinder ReponseFinder MySQL DB Stats Import Please select a data matrix for Hierarchical Cluster Analysis IEEE EDIDIT ETE REDE TEE EE TETTE EET EET ETE EE E ETE ERE TEE E E ESTEE PESE ETE TEE ERE TEE TEE EET EET EE EET ETE EE ETE ETE ERE E EET ITEDET Y GCMS IDBatchReport ll August2008 5 14 47 manMATRX 0 Ll M Clustering Mode
63. d against perfluorotributylamine PFTBA mass calibrant using the atune u autotune method provided with Agilent GC MSD Productivity ChemStation Software ver D 02 00 SP1 Agilent Technologies Palo Alto CA USA Product No G1701DA prior to the beginning of each 18 Data Acquisition analytical sequence Total run time was 58 5 min 19 20 DATA PROCESSING PARAMETERS 21 Field Value 22 Mass Spectral Extraction Software AMDIS Version 2 62 Raw GC MS data files were processed in AMDIS using the same AMDIS deconvolution settings as used in construction of the Golm Metabolome Database GMD MSRI library Q_MSRI ie adjacent peak subtraction 2 resolution low sensitivity very low medium shape requirements low described at http csbdb mpimp golm mpg de cgi bin madb2ml cgi inp m 2 amp c ml amp met all amp org msri amp lib all amp typ met amp o ht After deconvolution extracted spectra were identified either manually or by mass spectral matching to the GMD Q_MSRI library or NISTOS library Extracted spectra together with their AMDIS calculated Kovat s retention indices were then added to an AMDIS msl library This msl library was converted to the tabular MetabolomeExpress MSRI format using the MetabolomeExpress conversion tool Analyte specific quantifier ions were then selected for each library entry based on manual examination of raw signals in chromatogram regions centered around each analyte peak These quantifier ions were add
64. e Project Name bigger project of which this experiment is a part v wv 2 Give the full name of the Biological Experimentalist Name person who performed the biological part of the experiment Biological Experimentalist Email v vv 2 Give the name of the Metabolome Analyst Name person responsible for carrying out the metabolomic analysis Metabolome Analyst Email v vvn 10 Explain the hypothesis Experimental Hypothesis behind the experiment in at least 10 words v vwn 30 Describe in at least 30 words how the experiment was Brief Description of Experiment carried out including what other major pieces of data were acquired v vwn 5 If applicable provide a literature reference including full author list to the article where this experiment has been published v vwn 1 If applicable give the unabbreviated name of the journal in which this experiment has been published Publication Date YYYY MM DD v Jid PubMed ID v lint Literature Reference Journal You can see the validation codes in the Value column The different codes will be explained below 7 3 Validation Codes The general code format is as follows Validation level data type configuration parameters Instructions for completion optional Validation Level Validation level may be v or v v field value must be present and must match the assigned data type and configuration parameters v field value is optiona
65. e it has been backed up in a file with name ending in _METADATA_BACKUP TXT You should open it up in Excel and have a look You may want to add some details such as Genotype x Environment Class Comparisons or Instrument Batch IDs which may be important later on 1 5 Building and using custom Mass Spectral and Retention Index MSRI libraries 1 5 1 An overview of the MetabolomeExpress MSRI library format While MetabolomeExpress provides public MSRI libraries built under standard GC MS operating protocols many users will want to use their own GC MS methods and or MSRI libraries MetabolomeExpress uses a simple tab delimited format for MSRI libraries filename extension MSRI explained shortly and these may be uploaded into either the public or internal subfolder of the rares folder present in a user s FTP repository Libraries added to the public subfolder will be made publicly accessible to anonymous users via the MetabolomeExpress web interface Libraries added to the internal subfolder will only be accessible to users who are logged in and have permission to access that repository The MSRI format in its simplest form is a tab delimited text file table which may contain any number of columns provided it has a certain minimum set of 5 columns required by MetabolomeExpress The screenshot below shows an example of a library with the minimum set of columns ie Name HI Quantifier lons ID and Mass Spectrum
66. e Statistics tool 6 4 Comparative statistics This tool provides you with the ability to select a data matrix and then select some sample class comparisons to make by Welch s t test before being presented with an interactive heatmap of fold differences and p values The heatmap is linked to the raw data viewer in that if you double click on a cell containing a fold difference you will be automatically taken to the Raw Data Viewer and presented with the underlying GC MS signals This is great for building up confidence in particular results Here is an example Go to the Comparative Statistics tool select a data matrix and hit the Upload button Statistics and Data Exploration Control Panel Match Report to Data Matrix Data Matrix Renormalisation Matrix Explorer Comparative Statistics PCA HCA Correlation Analysis ResponseFinder ReponseFinder MySQL DB Stats Import Please select a data matrix for Welch s t test analysis l GCMS_ID Batch Report 20 August 2008 12 12 13 mzrtMATRIX v Statistical Options BT Maximum p value for highlighting of significant results 0 05 Only highlight results that are significant at the Bonferroni adjusted p value threshold EI Flag results based on skewed non normal data Maximum absolute skewness value 1 5 Wait a moment and you will be presented with a selection box asking you to choose your sample class comparisons of interest like this Genotype x Environment G x E Co
67. e control panel out of the way at this stage Chromatographic View EIC Overlay RT 0 to 60 min mz 147 9 00e 6 ey 30407 ANTIAOH MEOH OH 1 8 00e 6 7 00e 6 6 00e 6 n Intensity 5 00e 6 In 4 00e 6 3 00e 6 2 00e 6 1 00e 6 mn mam nm w o mn mn a 0 00e 0 Retention Time min 3 5 Zooming in and out Now in this case there are a lot of peaks making it hard to see what s going on To zoom in you must first select the region you wish to zoom in to by moving the pink selection window over it move the selection window by moving the mouse cursor over the chromatogram To resize the selection window click once resize to the desired width and then click again Then once you have the selection window covering the region of interest hold down the A key on your keyboard and double click Below is the result of Zooming in to the region of alpha ketoglutarate methoxime 2TMS which elutes at 23 86 minutes Remember to see the green peak annotation markers you must select the run for display of peak annotations by clicking on its name in the bottom part of the control panel Holding the mouse over green marker over the peak at 23 86 min shows that it was matched to alpha ketoglutarate methoxime 2TMS To zoom out again hold the Z key and double click The smaller your selection window the further you will Zoom out MetaData Viewer Raw Data Viewer Data Import and Peak Detection MSRI Library Matchin
68. e name of the RI hit in the RI Hits column is blue rather than black it has been positively matched to a library entry because one or more of the library specified quantifier ions was found indicated in red text and the MST ion ratios agreed with those in the library spectrum within the user specified tolerance parameters Below is an example of a nice clear positive match to Glycine 2TMS Cupar mre o 3287 HM ees vorn DD zm Greg 27148 ID 18 Daci bz 8 1 Ha Hao E Hu pteeteatat Good Good on Qual Qual Qual m z a al Qual p long Jone lons 1106 7 43 949 SIAT A eo eo oer ee CAE Daer EP Um app S08 MES 1a le TERRE eS E Ak MS a ET 8B 298 311 Quan ong depecped mz 204 147 102 W na EE n 21 H 3 amp 1 m wo mn pa i 1107 13 ET abt f 277 ort aM i 14 p tt Ghycin 2745 ID 18 Daci MS Oe 073 13 508 CR TET ER eM SINT Girdne 2 TMCS ID 18 Casin pg 08 1107 6 13312 Sep AES TS en Giycine TRES ICI 10 Danay MS 12 Dapary ese bpactrel Comprise WA 13318 aziz lat MEST GAMOT ARTA GF MEH OF 1 PEARLSI Ir 7 Deeg at 13 700 men 14 hati bud Se Dages i 1108 1345 Sam MA Bee na wi 17 ite Syne Satan b asd iy riu r i Sek il sit oie and nde c dure eg eg rt ST Ud abeo bau boo esu Lagu PRECES Se Ds L t Ges ag OO ox x me Vis RETI AE ATE Liwer Seeche Big Giyme TMS 38 R ae eg al dl 1 D st Ben Set gie Ww RT At acier giang T DI IE a hens pup AT 2f 46 qu pa
69. e riam al Dea ge id esch iim Thes barraat i M ID Beck Bap EX uel hea tres ra emperterd drente rin Agbesi l racer das Kr Akt Tuck ag Gare Spreng HE Age 3 27 mara DR Normalisation Overview caked GCM5 ID Batch Report 27 July 2008 13 27 43 nomnareaMATRTX You can download t usi O3040F AO0xX04UOI ADIOS4UI AU10407 AOIDAUPF A HTIAOUOHMENTIAOUHHMERTIAUOHMENTIAUOHMENHTIAUOHME OHOUHI PE OHOH PE OHUH3 PE OH UHA PE GQHOHS PE AKLIST AKRKLISTI AKLIST AKLIST AKLISI RIIO22 0 ID5 RI1022 8 MI314 Lactic acid 21145 He RT1028 1 Putative but Unconfirmeci ID6 RIIU28 1 Mri L Alsnines R trifluoroacetyl trineethyksbyl ester SE RILUAA 5 DD RIIO44 6 HZ2 Match cur If you have a valid metadata file in the experiment folder the result will look more like this metavoromne Havigatian H Database rapiti A jPlantnargykisisy meng sE Pineng Bi Public a Arabidopsis ACA KO Moderate High Ligh a jAabidapsis Cell Suspension Timecoure p aj J arabidepsis GST AHAJ S4 and HzO2 Tres E CD Arabidopsis Mitochanerial ETC Carmela i H ai Arabidopsis Mipchandrial ETC Corian H a Arabidopels SAL 1 Mutants abii and iry1 1 zi 3 Example Exparenen Arasidopsis Gels An T 30407 AMTIAOH MEGH OH 1 Cp Ej 090407_ANTIAOH MEOH 0H VLC SS 030407 ANTIA OH MEOH OH 2 COF ag 37 030407 ANTIA OH MEOH OH 3 ME 8 5 Si 3 E 5 030407 ANHTIA GH MEOH DH mF 030407 AHTIA GH MEOH DH APE
70. ed as a MATCHREPORT file in the experimental folder on the server you will just be able to review the library matching results in the Output window The process used by the MSRI library matching algorithm is best described using the following decision tree Mass Spectral Tag MST reconstruction and the MSRI library matching process Retention indices of all EIC peaks calculated relative to internal or external RI calibrant peaks by linear interpolation All EIC peaks assigned to 0 1 RI unit bins Mass Spectral Tags MSRI library matching procedure Step 1 Is Mass Spectral Tag MST within RI tolerance window of an MSRI entry IF YES Go to Step 2 IF NO Move on to next MST and begin at Step 1 Step 2 Does the MST contain any of the MSRI library specified quantifier ions IF YES Gather the m z intensity pairs from all MSTs within the user specified MST centroid distance 1 0 RI Units by default and merge these temporarily with the current MST Count the number of other ions in the merged MST that are within a set percentage of the expected intensity based on the intensity of the quantifier ion and the full mass spectrum in the MSRI library Calculate the average deviation of all ions from their expected intensities Move on to Step 3 Repeat for each detected quantifier ion IF NO Repeat Step 2 for any remaining RI matches in the MSRI library If no more RI matches remain move on to next MST and begin at Ste
71. ed to the Quantifier lons column of the MSRI 23 Mass Spectral Extraction Procedure table 24 25 MSRI Library Entries 26 Name RI RT 27 L 2 Aminobutyric acid 1TMS 993 3 28 l Trimethylsiloxy 2 trimethylsilylaminc 994 29 Pyridine 3 trimethylsiloxy MS80 RI1 1000 4 wima 1TNAACY AACOT an 20 Duende acid matha Md b gt CPEB STANDARD MSRI J Note above how there is a blank line and then the line MSRI Library Entries before the main table starts 1 5 4 Displaying and Validating MSRI Libraries with MSRI Library Manager The contents of an existing MSHI library may be viewed at any time by right clicking on the library in the MSAI Library Manager and selecting Display and Validate This will display any metadata in the library and generate a validation and review table as shown in the earlier screenshot It will also provide you with a hyperlink to the MSRI library file and automatically generate a template analyte annotations table which annotates each library entry with its matched standard underivatised metabolite name its InChl structure code and its chemical class These tables also provide four boolean ie TRUE 1 or FALSE 0 columns that allow you to specify whether each library entry is 1 to be used as a quantifier peak for its corresponding metabolite 2 of unknown structure 3 an internal standard or 4 an artefact analyte of non biological origin These tables are important for proper filtering
72. ee Section 1 3 below for details 1 3 Obtaining your own MetabolomeExpress data repository Registration In order to analyse and share your own GC MS datasets with MetabolomeExpress you must first register to obtain a username and password Registration is FREE Once your application to register has been received you will be contacted by email to determine whether you wish to create a new repository and if so what you want to call the repository or simply to join an existing repository If you wish to join an existing repository we will require authorisation via email from the original creator of that repository It IS possible for you to have your own repository and also have access to another repository via the web interface we just need to receive an email request to give you repository access permissions from the owner of the other repository NOTE Each registered username will obtain FTP access to only one repository If you wish to upload data to another repository you must obtain the username and password from an FTP user of that repository or register for another username and password for use with that FTP repository 1 4 Uploading and managing your datasets via FTP 1 4 1 Logging in To upload and manage your datasets via FTP you will need an FTP client program We use the free FTP client FileZilla but most FTP clients should work fine To connect to your FTP repository enter the following into your FTP client pro
73. efining core metabolic and transcriptomic responses to oxygen Rice germination timecourse germination in air vs germination in Rice WT seedling Germ N2 24h Rice WT seedling 3 availability in rice embryos and young seedlings Plant Physiology Epub ahead of print N2 Germ Air 24h Howell KA Narsai R Carroll A Ivanova A Lohse M Usadel B Millar AH Whelan J 2009 Mapping metabolic and transcript temporal ES wo e e M switches during germination in rice highlights specific transcription factors and the role of RNA instability in the germination process Rice aerobic germination timecourse Rice WT miadiing Gem Aic 7h Rice WT send 10 EE Dry Seed Oh Plant Physiology 149 2 961 80 Howell KA Narsai R Carroll A Ivanova A Lohse M Usadel B Millar AH Whelan J 2009 Mapping metabolic and transcript temporal ARIS E z Ss gt e switches during germination in rice highlights specific transcription factors and the role of RNA instability in the germination process Rice aerobic germination timecourse Rice WT seedling Germ Air 48h Rice WT seed 9 i 3 Dry Seed Oh Plant Physiology 149 2 961 80 Wilson PB Estavillo GM Field KJ Pornsiriwong W Carroll AJ Howell KA Woo NS Lake JA Smith SM Harvey Millar A von Yes Te S J Caemmerer S Pogson BJ 2009 The nucleotidase phosphatase SAL1 is a negative regulator of drought tolerance in Arabidopsis Plant Metabolomic character n of drought tolerant a8 and bt alx8 Leaf NormalGrowt
74. ens Root Control 5h zl Populus x canescens Root Flooding 24h Populus x canescens Root Control 24h E Populus x canescens Root Flooding 168h Populus x canescens Root Control _168h 4 d d Refresh Gm Clicking GO generates the following result in the MetaAnalyser display only the top few results are visible fib Home A Experiment Explorer ijg Database Explorer Hep Z Registration Database Statistics ResponseFinder MetaAnalyser MSRI Library Manager MetaAnalyser Display MetaAnalyser Results Heatmap EEN The selected metabolite responses have been clustered by HCA You can download the clustering heatmap dendrogram results in PDF format using this link Only metabolites detected in all selected experiments were included You may sort the table below by clicking on the table headers NOTES The values in the table below are the signal intensity ratios SIRs between the experimental classes indicated in each header For an SIR gt 1 ResponseValue SRR 1 For an SIR lt 1 ResponseValue 1 SIR 1 Responsiveness Absolute Value ResponseValue Average Responsiveness the average Responsiveness of a metabolite across all selected class comparisons Standard Deviation the standard deviation of all ResponseValues of a metabolite across all selected class comparisons Kreuzwieser J Hauberg J Kreuzwieser J Hauberg J Kreuzwieser J Hauberg J Howell KA Carroll
75. er cell line Normoxia cultured cell 1 100 25 151009 cancer cells normoxia Rep 6 cancer cell line Normoxia cultured cell 1 100 sometimes you may be interested in comparing the metabolomes of different parts of an organism under different conditions For example the screenshot below shows a MINIMET TXT file for a hypothetical experiment comparing the metabolomes of plant roots and shoots under normoxic and hypoxic conditions keng A B C D en Organ or Sample Sample ID Genotype Treatment Biomaterial Timepoint Mass or i Type Volume 2 151009 shoot hypoxia Rep 1 wildtype Hypoxia shoot 1 100 3 151009 shoot hypoxia Rep 2 wildtype Hypoxia shoot 1 100 4 151009 shoot hypoxia Rep 3 wildtype Hypoxia shoot 1 100 SS 151009 shoot hypoxia Rep 4 wildtype Hypoxia shoot 1 100 6 151009 shoot hypoxia Rep 5 wildtype Hypoxia shoot 1 100 7 151009 shoot hypoxia Rep 6 wildtype Hypoxia shoot 1 100 8 151009 shoot normoxia Rep 1 wildtype Normoxia shoot 1 100 9 151009 shoot normoxia Rep 2 wildtype Normoxia shoot 1 100 10 151009 shoot normoxia Rep 3 wildtype Normoxia shoot il 100 11 151009 shoot normoxia Rep 4 wildtype Normoxia shoot 1 100 12 151009 shoot normoxia Rep 5 wildtype Normoxia shoot 1 100 13 151009 shoot normoxia Rep 6 wildtype Normoxia shoot 1 100 14 151009_root_hypoxia_Rep 1 wildtype Hypoxia root 1 100 15 151009 root hypoxia Rep 2 wildtype Hypoxia root 1 100 16 151009 root hypoxia Rep 3 wildtype Hypoxia root 1 100 17 151009 root hypoxia Rep 4 wi
76. ess Metadata Validation Templates 7 1 Background oystematic annotation of biological datasets with contextual metadata adds great value to primary data by enabling the systematic analysis of relationships between contextual variables technical parameters and biological phenotypes Systematic annotation combines the use of defined data structures and controlled vocabularies or ontologies to facilitate computer based processing of annotated data In MetabolomeExpress annotation of datasets is achieved through the use of tab delimited metadata METADATA TXT files Each experimental dataset in MetabolomeExpress is accompanied by a single metadata file providing answers to the following questions e Who did the experiment and how to contact them e What was the aim of the experiment e What were the genetic and phenotypic characteristics of the organisms studied e What environmental conditions were the organisms exposed to e What perturbations were applied to the organisms e What parts of the organisms were analysed e How were samples from the organisms processed and analysed The specific types of contextual information pertinent to a particular experiment and the range of possible values for each variable depend on the type of organism studied the type of environment they were studied in and the type of experiment carried out with them Minimum field sets appropriate to each research field are being determined by the Metabolomic
77. etting started 1 1 Structural overview MetabolomeExpress is comprised of three interacting layers 1 An FTP repository where registered users may upload and manage their own GC MS datasets 2 A MySQL database that stores general metabolite information and metabolite response statistics from datasets present in the quality controlled MetabolomeExpress database of metabolite response statistics 3 A web interface to interact with data present in the FTP repository and MySQL database A FTP DATA REPOSITORY FTP Repository f Repository 1 eg PlantEnergyBiology libraries public Public MSRI Library 1 MSRI Public MSRI Library 2 MSRI internal E Private MSRI Library 1 MSRI Private MSRI Library 2 MSRI I Private Folder eg PlantEnergyBiology Internal H Private Experiment 1 Metadata File METADATA TXT I NetCDF Raw GC MS File s CDF eXtracted lon Chromatogram File s SICH Peak Detection Result File s PEAKLIST AMDIS RI Calibration File s CAL Library Matching Report File s MATCHREPORT Data Matrices mzriMATRIX Statistical Processing Results various formats Private Experiment 2 Public Folder eg PlantEnergyBiology Public Public Experiment 1 I Public Experiment 2 B SERVER SIDE PROCESSING MODULES WEB INTERFACE a Home Basic info Public announcements A Experiment Explorer Meta Data Viewer R
78. fer line temperature between the gas chromatograph and mass spectrometer was 300 degrees C Electron Impact El with 70 eV ionisation energy and ion source temperature of 230 degrees C Quadrupole temperature was set to 150 degrees C After a 10 minute solvent delay filament 1 was turned on and full scan spectra acquired in the range m z 40 600 at a rate of 2 6 spectra s The mass selective detector was pre tuned against perfluorotributylamine PFTBA mass calibrant using the atune u autotune method provided with Agilent GC MSD Productivity ChemStation Software ver D 02 00 SP1 Agilent Technologies Palo Alto CA USA Product No G1701DA prior to the beginning of each analytical sequence Total run time was 58 5 min Cy dW 9 CG 7 CPEB STANDARD MSRLMSRI Microsoft Excel IM X a Home Insert Page Layout Formulas Data Review View Add Ins Q x m wx EES a Insert E B ce u qw mm ER ee em gt Y EE as 3 Delete Al Paste c else se o 0 00 Conditional Format as Cell Sot amp Find amp B I U i O A BE B Ble HECKER SEA 8 398 L MI i 7 Formatting Table Styles BH Format 4 Filter Select Clipboard Font E Alignment E Number E Styles Cells Editing C51 v 3 E E R A B After a 10 minute solvent delay filament 1 was turned on and full scan spectra acquired in the range m z 40 600 at a rate of 2 6 spectra s The mass selective detector was pre tune
79. g Statistics and Data Exploration Help Chromatographic View 1 EIC Overlay RT 23 438 to 24 227 min mz 147 2 0e 5 wY v 030407 ANTIA OH MEOH OH 1 n ji Peak Info Run ID 030407_ANTIA 0H MEOH 0H 1 I zd RT 23 861 e I A RE 1576 9 ai ea RI Observed Expected 1 1 2 WW E Quantifier m z 288 C CG Analyte 2 Ketoglutaric acid methoxime 2ZTMS e MS Match 5 14 35 7 Good ierg Avg MS Ratio Error 36 9 o l 1 0e 5 Ij n A A l N A j Vh f KS N ul A d V 0 0e 1 CC EE m rs s lol LLLI LLLLIIJBLLILLLLLJ 23 4 23 5 23 6 23 7 23 8 23 9 24 24 3 Retention Time min 3 6 Viewing mass spectral scans If you find an interesting peak you can check out the mass spectral signal captured at that retention time by moving the left hand edge of the selection window over that retention time holding down the SHIFT key and double clicking The mass spectral scan will then be displayed in the Mass Spectral View below the Chromatographic View Below is the spectrum observed at the apex retention time of the peak matched to alpha ketoglutarate methoxime 2TMS Mass Spectral View 030407 ANTIA OH MEOH OH 1 3 7 Finding differentially expressed peaks using the chromatographic statistical tool To quickly find biologically intere
80. gram Hostname www metabolome express org username your username password your password and connect Once connected you should see your repository This screenshot below shows an example of the result of logging in as AdamC using FileZilla fr AdamC www metabolome express org FileZilla Fe Edt View Transfer Server Bookmarks Mep New version avakibiet OG BO BAUO8R tui Host www metabolcm Username Adamc Password LI 227 Entering Passive Mode 150 203 72 67 118 152 LIST 150 Here comes the directory listing 226 Directory send OX Directory listing successful e c3 Data Downloaded from SetupX amp O Etienne Complex I Mutants 300608 C3 Ne s Quant Proteomics Data LC Q TOF it O Permis GCMS data Q QTOF Training Data Rice Germination Timecourse Ar vs N2 fiese Fletype Lastmo Permes O Fie Foder 2009 10 11 dewxr xr x 48 Fie Folder 2009 10 06 drwxr xr x 48 Fie Fokler 2009 06 27 drwxr xr x 48 Queued files Foiled renders Successful transiers Z Queue empty ee Once connected there are a few simple steps to create a new experiment Step 1 Create a data folder for the experiment Open either your public or internal data folder and create a new subfolder with the name of the folder being the name of the experiment eg Nutrient supplementation timecourse 1 Step 2 Upload raw GC MS files Copy all the raw NetCDF ANDI MS AIA GC
81. h Col 0 Leaf NormalGrowth 8 eris r Arabidopsis mutants Journal 58 2 299 317 Garmier M Carroll AJ Delannoy E Vallet C Day DA Small ID Millar AH 2008 Complex I dysfunction redirects cellular and Timecourse Treatment of Arabidopsis cell suspension cultures with At Ler Cell Suspension Cell suspension Rotenone_3h At 8 mitochondrial metabolism in Arabidopsis Plant Physiology 148 3 1324 41 rotenone Ler Cell Suspension Cell suspension Methanol_3h Narsai R Howell KA Carroll A Ivanova A Millar AH Whelan J 2009 Defining core metabolic and transcriptomic responses to oxygen x i pr E Rice WT seedling N2 ir Switch 25h Rice WT seedling i availability in rice embryos and young seedlings Plant Physiology Epub ahead of print Rice germination timecourse N2 Air gas switch N2 EA w t P 03 2 ars M in sei Retrieved metabolite responses receive 1 point towards their Phenocopy Score for each metabolite that responds significantly p 0 05 in the same direction in both bait and tested response For example if alanine is significantly increased in the bait and it is also increased in a response in the database then that response will receive 1 point The more metabolites that respond significantly in the same direction as in the bait response the more points a retrieved response will have High Phenocopy Scores may indicate that the bait and retrieved responses share a common underlying mechanism This notion
82. have not had there structures verified by comparison with authentic standards You will then be given a chance to select which samples to include in the PCA By default all variables are scaled to unit variance before the PCA is carried out For this example you don t need to change anything just hit the Upload button Statistics and Data Exploration Control Panel ES Match Report to Data Matrix Data Matrix Renormalisation Matrix Explorer Comparative Statistics PCA HCA Correlation Analysis ResponseFinder ReponseFinder MySOL DB Stats Import Please select a data matrix for Principal Components Analysis GCMS_ID Batch Report 11 August 2008 5 14 47 mzrttMATRIX v Filter redundant analytes Remove analytes of unknown chemical structure Refresh You will then be given a chance to select which samples you wish to include in the PCA Sample Selection Please select the samples to be included in the PCA 050407 ANTIA 6H MEOH 6H 1 PEAKLIST 050407 ANTIA 6H MEOH 6H 2 PEAKLIS T 050407 ANTIA 6H MEOH 6H 3 PEAKLIST 050407 ANTIA 6H MEOH 6H 4 PEAKLIST 050407 ANTIA 6H MEOH 6H 5 PEAKLIST 050407 METHANOL 6H 1 PEAKLIST 050407 METHANOL 6H 2 PEAKLIST 050407 METHANOL 6H 3 PEAKLIST 050407 METHANOL 6H 4 PEAKLIST 050407 METHANOL 6H 5 PEAKLIST For example try selecting the 6 hour Antimycin A and Methanol treated samples and then hit Submit Wait a moment and the following items will appear in the Output windo
83. he table The other way is to add a metadata section to the top of the library file Adding metadata about how the library was made is essential if it is to be publicy disemminated To do this add the following line to the top of the file MSRI Library Attributes You may then add three metadata sections to the file ie ADMINISTRATION INSTRUMENTAL PARAMETERS and DATA PROCESSING PARAMETERS The following screenshot shows how these sections are added to the file Essential lines and cells are highlighted with a light blue background and bold text this is for illustration purposes only no formatting is stored in the tab delimited file The fields shown below are recommended but the actual field names and their values are totally flexible and you may add as many fields to each section as you like MSRI Library Attributes ADMINISTRATION Field Library Builder Name Library Builder Email 6 Library Description 8 INSTRUMENTAL PARAMETERS CPEB STANDARD MSRLMSRI Microsoft Excel Conditional Formatas Cell Formatting Table Styles Dr Adam James Carroll adam carroll anu edu au This GC MS mass spectral and retention index library was built from raw GC MS data produced by the analysis of a variety of complex methanolic plant extracts and authentic metabolite standards derivatised by methoximation and trimethylsilylation Analysis was conducted on an Agilent 6890N GC fitted with an Agilent 5975B
84. ibrary matching and quantitation 030407 ANTIA 0H MEOH OH 1 PEAKLIST 030407 ANTIA OH MEOH OH 2 PEAKLIST 030407 ANTIA OH MEOH GH 3 PEAKLIST 030407 ANTIA OH MEOH GH 4 PEAKLIST 030407 ANTIA OH MEOH DH 5 PEAKLIST Matching Criteria RI window MST centroid distance 1 0 Lut Rr unis Min Peak Area for peak import 2500 Arbitrary Peak Area Units MS Qualifier Ion Ratio Error Tolerance a0 Min Number of Correct Ratio MS Qualifier Ions 2 Max Average MS Ratio Error op MSRI CPEB STANDARD MSRI MSRI MERI Library Manager RI Calibration Fle Alkane RI 030407 Ohel O o Om Use RI calibration file specified in metadata file amy selection above will be ignored Carry out per sample fine calibration using internal RI standards Internal RI Standard Specific Marker Ion m z Output Options Display Library Information Table F dese Results erwes Display MSTs with no RI match Display MSTs with RI match s but no detected quantifier ion s du Administrator User Name Administrator Password Oo o 1 A progress bar should appear in the lower part of the screen Wait until progress is complete and in a few moments you should be presented with a large interactive report table in the output window like shown below you may wish to collapse the control panel at this point to get it out of the way Deapen atn MSRI Library Search Results Been nien E e Detecled Zeechen cols RI Hil M3 Malch Det
85. idity Temperature Watering Regime Nutritional Regime Date of Establishment Environment 2 TREATMENTS Treatment 1 Treatment IO Treatment Group ID Treatment Class Biotic Abiatic Chemical Mone Treatment Description Treatment Dose Treatment Dose Units Treatment Date VYY MBSI DDJ Treatment Time HH amp Thl 55 Treatment Duratian Treatment Duration Units Treatment 2 GxE COMPARISONS GxE Comparison 1 Gk Comparison IL Genotype ID Mumerat r Treatment IC Numerator Organ Numerator Genotype ID Denominator Treatment ID Denominatar Organ Denmnominatar GxE Comparison 2 DT COLLECTIONS Collection 1 Colkection ID Harvest Date vry MM DT Harvest Time HH h4M 55 Developmental Stage Developmental Stage Scale Organ Organ CV Metabolism Quenching Methed Harvest Method Harvest Amount Storage Collection D I i E d D EXPLANTS explant 1 Explant ID Genotype ID Environment ID Treatment ID Mane Collection ID Explant 2 AL ANALYSIS META BIOSAMPLE PROCESSING PROTOCOLS BioSample Processing Protocol 1 Bioc5armnle Processing Protepol IE Bic5ample Processing Protocal Description Sample Storage BioSample Processing Protocol 2 BIOSAMPLE EXTRACTION PROTOCOLS BioSample Processing Protocol 1 Extraction Protool 1D Extracrien Protocol Descriptio Ewiract Concentration Extract Cigar tn Extract Storage BioSample Processing Prot
86. ik Are Fee tae epe 23000 miy Pea Ama Lem MS Cecile Does Kee Eri Toiegeg j Mic amber c Corect Baro M5 Goler fone 2 Man Average M Rann Eman 4 L MSE Lire CPER STANDARD MSp MSp ab e Baca Matt Sqecrral Comparison Ri Calbeasoo Pik Scans BI O30207 Ce x Uns RI iser Fin specie in ed m ay icto above oil be geg Curry Du Dit dur pia Deg DEFIO vq internal EI AiE wl Esita AE Starsdacd Esc Hehe Des es B5 Output prions Sitem Statut Dhea L t ELOA ei zl Dar Pati Lien plan MT wih ns AS manes Crac MEST a with RI pa but no Geen quanta ient d Agen Patel To initiate a library matching process you use the control panel to select one or more PEAKLIST files for library matching set library matching criteria choose an appropriate MSRI library and RI calibration file soecify whether to carry out fine per sample RI calibration by the finding of internal RI standard peaks specify an appropriately characteristic ion of the internal RI standards m z 85 is good for commonly used n alkanes specify a number of output options and hit the GO button You will need to be logged in and have write permissions on the relevant repository if you wish to process more than one sample and generate a MATCHREPORT file in the experiment folder If you aren t logged in or don t have write permission you can still carry out library matching but only for a single file the first selected file in the list and your results won t be stor
87. include Data File The name of the PEAKLIST file without path information representing the sample in that column Tissue Mass Volume The mass or volume of tissue fluid that was extracted to produce the analytical sample Genotype ID The Genotype ID of the organism that was analysed as given inthe METADATA TXT file for the experiment Organ The standard name of the organ tissue biomaterial type that was taken from the organism and processed to produce the sample Treatment ID The Treatment ID of the experimental treatment applied to organism that was analysed Treatment Duration The treatment duration applied to the organism that was analysed Treatment Dosage The treatment dosage applied to the organism that was analysed Replicate The replicate number of the sample For a given genotype treatment organ combination replicates are numbered 1 x where x is the number of replicates How header columns include Analyte Signal ID The library match annotation as given in the Library Hit Name column of the MSHI Library Match Report It must be given in the following syntax variables displayed in italics constants displayed in bold Name of Library Entry IDID of Library Entry RIRI of Library Entry MZnvz of Quantifier lon Average Retention Time min The average retention time of the matched quantifier ion signal across the entire row of the table Average Retention Index Kovats The average Kovats retention index
88. ion Protocol ID Example Valid Value Any value matching one of the values given for Extraction Protocol IDinthe BIOSAMPLE EXTRACTION PROTOCOLS metadata section Experimental File Reference fileref exp Data Type Code fileref exp Validation Description The value must be the name of a computer file without any path directory information that exists in the data folder of the experiment not in a subfolder Case sensitive Configuration Parameters None Example Code v fileref exp Example Valid Value calibration data cal would pass validation if the file calibration data cal was found in the same experimental data folder as the metadata file being validated Instrumental Data File Reference run ref Data Type Code run ref Validation Description The value must be the name of a NetCDF CDF file without the CDF file name extension or any path directory information that exists in the data folder of the experiment not in a subfolder Case sensitive A PEAKLIST file and an XIC file with the same name except with a PEAKLIST and XIC file name extension respectively instead of CDF must also be present in order to pass validation Configuration Parameters None Example Code v run_ref Example Valid Value mutant1_rep1 would pass validation if the files mutant1 rep1 CDF mutant1_rep1 XIC mutant1 rep1 PEAKLIST were all found in the same experimental data folder as the metadata file being
89. ire eoa Sir eg Macr KT r Eh binds bene Sommerer Fee EE Fees Mega SC mdi se frt Deg 6 WE n mi bie am You can see the MST containing the library specified quantifier ions labelled with Quant ion s detected m z 204 147 102 If you click the Display MS buttons next to the list of the MSTs mass intensity data and the name of the RI Hit you can display the MST spectrum and library spectrum respectively in the Mass Spectral Comparison window as shown in the screenshot above 6 Statistics and data exploration Click on the tab entitled Statistics and Data Exploration and we can go through the various statistical tools Remember to use these tools you must have already loaded a dataset as described earlier Statistics and Data Exploration Help This is where you construct data matrices from MSRI library matching reports and then explore the matrices using an array of easy to use univariate and multivariate statistical tools You will then see a screen like this db dl w L http ecabost MetabolomeExpress metabolome xpr Mavwsgatezn e Database May Output 3 SO CPS Uezabokesefaprean Server y Lj Example Experiment Aratetapsas Cel Suspen 5S S IHE ONLINE TOOLKIT FOR HIGH THROUGHPUT GC AMS METABOLOMICS DATA ANALYSIS AND SHARING Lira insa Law Data Wiewer Cute imported Peak Detecben BERI Liroy Matching Statisties and Daka Exploration Hele Statistica ard Data Exploration Contes
90. is supported by the fact that the top scoring hits here are the responses to rotenone at different time points and a response of rice seedlings to anaerobic germination compared to aerobic germination 3 Processing and analysis of experimental GC MS datasets with Experiment Explorer 3 1 Example dataset Timecourse metabolomic analysis of plant cells treated with Antimycin A The easiest way to learn is by example So here we will provide a step by step guide that will show you how to start exploring the example GC MS metabolomics dataset featured in the MetabolomeExpress publication a 24 hour timecourse analysis of Arabidopsis thaliana plant cells responding to pharmacological inhibition of the mitochondrial electron transport chain with the classic respiratory inhibitor Antimycin A A brief description of the example experiment At the beginning of the experiment a number of 120 ml cell suspension cultures were sampled and immediately treated with either 25 uM Antimycin A final concentration supplied as a 100 ul dose suspended in methanol methanol 100 ul or water 100 ul The cultures were then re sampled after 1 3 6 12 16 and 24 hours of treatment Hence there are a total of 21 different treatments in the dataset 3 treatment groups each including 7 different treatment durations Given that each treatment was replicated 5 times that works out to about 100 individual GC MS runs or about 5 6 GB of raw GC MS data Now that s e
91. ith the same filename as the NetCDF file from which it was derived eg the peak list file corresponding to the NetCDF file called 20091510 Wildtype 1 CDF must be named 20091510 Wildtype 1 PEAKLIST 1 6 4 MSRI library matching report tables MATCHREPORT MetabolomeE xpress library match report tables are used to annotate chromatograms in the Raw Data Viewer and also to construct data matrices using the Match Report to Data Matrix tool of the Statistics and Data Exploration component of the Experiment Explorer The only requirement for the naming of library match reports is that they end with the file extension MATCHREPORT This enables them to be recognised as library match reports by MetabolomeExpress The file format for library match reports is a tab delimited table with the following columns each row representing a single MSRI library match Datafile the name of the peak list file that was processed to generate this library match without any path information Library Hit Name The name of the library hit The MetabolomeExpress MSRI Library Matching algorithm outputs the library hit name in the following syntax variables displayed in italics constants displayed in bold Name of Library Entry_ IDID of Library Entry RIRI of Library Entry MZnvz of Quantifier lon NOTE This syntax is important SO be sure to use it if building your own library match reports using third party software Intensity The total peak area of the quan
92. ith value of either 1 TRUE or 0 FALSE Set to 1 if the analyte is of unknown structure or has not been verified with authentic standards Library entries not automatically matched to known metabolites will be automatically set to 1 is quant peak This is a boolean with value of either 1 TRUE or 0 FALSE Set to 1 if you wish peak areas for this library entry to be considered as representative of levels of its corresponding metabolite Set to 0 if you want to exclude values for this peak from quantitative analyses his is useful for preventing highly variable or unreliable metabolite derivatives from influencing results is artefact This is a boolean with value of either 1 TRUE or 0 FALSE Set to 1 if the library entry represents an analytical artefact analyte of substantially non biological origin Otherwise set to O Quantitative data thus annotated as corresponding to artefacts will be automatically removed from data matrices prior to multivariate analysis is internal standard This is a boolean with value of either 1 TRUE or O FALSE Set to 1 if the library entry represents an internal standard analyte of non biological origin eg n alkanes FAMES ribitol Otherwise set to 0 Quantitative data thus annotated as corresponding to internal standards will be automatically removed from data matrices prior to multivariate analysis The screenshot below shows some example rows for different types of analytes A B C D
93. k intensity and an appropriate biomass volume correction factor The data matrix renormalisation tool let s you automatically renormalise mzrtMA TRIX format data matrices using metadata stored in the tab delimited metadata file kept in the experiment folder with all the other raw and processed experimental data Currently there is only one type of renormalisation available but more will be added in the future The currently available method is to normalise each metabolite abundance value to the mean of its experimental control values Appropriate controls are defined in the metadata file and cannot be changed except by editing that file If you want to see which sample classes are normalised to which other sample classes take a look at the table under the heading GxE COMPARISONS When you renormalise a matrix any value belonging to a sample class defined by the numerator parameters genotype organ and treatment will be divided by the mean of values from samples of the class defined by the associated denominator parameters genotype organ and treatment If you wish to experiment with your own methods of renormalising raw mzrtMATRIX data matrices using local software you can load a matrix with the Matrix Explorer and download it as a spreadsheet readable tab delimited text file using the provided hyperlink Anyway set the Data Matrix Renormalisation control panel up as shown below and hit the GO button Statistics and Data Exploration Cont
94. l It may be blank but if a value is present it must match the assigned data type and configuration parameters If neither of these validation level codes is present in the template field then no validation of values for this field will occur 7 4 Data types and associated configuration parameters Float f Data Type Code Validation Description Must be a number and the number may have a decimal point Configuration Parameters None Code Example v f Example Valid Value 1002 3652 Example Invalid Value 12defg Integer int Data Type Code int Validation Description Must be an integer Configuration Parameters None Code Example v int Example Valid Value 1 Example Invalid Value 1 654 Date d Data Type Code d Validation Description A date in YY Y Y MM DD format Configuration Parameters None Example Code v d Example Valid Value 2010 01 01 Example Invalid Value Jan 01 2010 Time t Data Type Code t Validation Description A 24 h clock time in hh mm ss format Configuration Parameters None Example Code v t Example Valid Value 11 59 59 Example Invalid Value 30 99 76 Date Time dt Data Type Code at Validation Description A date time in YY YY MM DD hh mm ss format Configuration Parameters None Example Code v dt Example Valid Value 2010 01 01 11 59 59 Latitude and Longitude latlong Data Type Code latlong Validation Description A latitude and longitude i
95. ldtype Hypoxia root 1 100 18 151009 root hypoxia Rep 5 wildtype Hypoxia root 1 100 19 151009 root hypoxia Rep 6 wildtype Hypoxia root 1 100 20 151009 root normoxia Rep 1 wildtype Normoxia root 1 100 21 151009 root normoxia Rep 2 wildtype Normoxia root 1 100 22 151009 root normoxia Rep 3 wildtype Normoxia root 1 100 23 151009 root normoxia Rep 4 wildtype Normoxia root 1 100 24 151009 root normoxia Rep 5 wildtype Normoxia root 1 100 E 151009 root normoxia Rep 6 wildtype Normoxia root 1 100 Once you have completed the MINIMET TXT file save it as a tab delimited file and upload it into the experiment folder by FTP Step 5 Upload a retention index calibration file The next data processing step after data import and peak detection is MSRI library matching However to perform MSRI library matching MetabolomeExpress requires a small retention index calibration file to be present in the experiment FTP folder The format used for this is the AMDIS CAL format This is just a small tab delimited text file providing the retention times and Kovats retention indices of a set of retention index RI calibration compounds eg alkanes spanning the retention time range of the used GC MS method You can either add these compounds to your samples prior to analysis highly recommended or analyse them in a separate run added to the same instrument batch sequence as your actual biological samples either way you will need to determine the retention
96. limited STATS File with redundant signals GCMS ID Batch Report 10 Sentember 2008 4 45 58 stats 16 Oclober 2008 13 28 54 REDUNDANT STATS Tab delimited STATS File non redundant GCMS ID Raich Report 10 Seotember 2008 4 45 58 16 Ociober 2008 12 20 24 NONREDUNDANT STATS Tab delimited Means and Standard Devistions Report with redundant signals GCMS ID Batch Report 10 September 2000 4 42 2016 October 2000 12 20 24 stats MEAN SD oq HTML Heatmap Table with redundant signals GCUS ID Batch Report 10 September 2000 4 45 58 stats 16 October 2000 12 20 54 himi HTML Heatmap Table non redundant organised by chemical dass GCMS Q wer 2008 4 45 5316 October 2008 13 38 54 L Tab delimited Cytoscape Format non redundant organised by chemical dass GCA NOTE The Bonferroni corrected p value threshold ie the user specified p value cutoff divided by the number of signals tested is 0 00018518518518519 HINT You may click on column headers to sort the table according to any analyte parameter Also double clicking on signal intensity ratio cells will take you to the underlying raw signals in the raw data viewer t Ler Cell At Ler Cell 1 Ler Cell t Ler Cell Suspension Ler Cell Suspensicn t Ler Cell Suspension Suspension Cuftured Suspension Cultured Suspension CuRured Cultured ceii cultured cell Cultured Co ell AntimycinA th Srel An mycinA 3h Arel Antimycina 6h A Afen 12h At AntimycinA 16h At A
97. lume This is achieved by dividing the peak area values in each column ie each sample by the internal standard peak area measured in that sample and then dividing that result by the mass or volume of biological sample that was extracted to produce the sample The MetabolomeExpress Match Report to Data Matrix tool currently provides two different internal standard normalisation options One approach is to normalise to a single internal standard that is added to each extract at some known constant concentration You can specify which signal in the data represents the internal standard by entering a unique identifier string that is present in the name of internal standard signal in your library matching results For example there is only one library entry called Ribitol in the CPEB STANDARD MSRI MSRI library used to process the example dataset so using the identifier string Ribitol will only match the Ribitol internal standard peak Normally pipetting errors are relatively small and the internal standard peak area in each sample should be pretty much the same A large deviation of an internal standard peak area from the median internal standard peak area is therefore a good indication that something more sinister than small pipetting errors has gone wrong and the data should therefore not be trusted Similarly if the internal standard peak is unusually small or cannot be found at all alarm bells should ring The MetabolomeExpress Match
98. mas ID A unique ID number or string for the library entry It is recommended that you keep these as short as possible because they will be included in library match annotations and displayed onscreen Mass Spectrum The mass spectrum of the analyte encoded as a series of m z intensity pairs where each m z intensity pair is given as the m z followed by a space and then the intensity followed by a semicolon and then optionally a space 1 5 2 Creating MSRI libraries from AMDIS MSL files How you build your library table is a matter of personal preference However we use the freely available deconvolution tool AMDIS to build libraries from reference chromatograms and then use the MetabolomeExpress MSHI Library Manager to convert AMDIS MSL format libraries to MetabolomeExpress MSHI format libraries and then use a spreadsheet to fill in the quantifier ion column It is also possible to convert MSRI format libraries back into AMDIS MSL format libraries To convert an AMDIS MSL format library into an MSRI format library upload the MSL file into your internal libraries folder and open the MetabolomeExpress MSRI Library Manager in your browser by navigating to Database Explorer gt MSRI Library Manager in the MetabolomeExpress web interface You will need to be logged in to see the libraries in your internal libraries folder Expand your In house MSHI Libraries Folder in the control panel gt right click on your MSL library gt
99. metadata file passes its corresponding validation test you will need to have read the section in this manual on interpreting validation templates You can download the latest validation templates from the Database Explorer module The core format structure of the file allows a data file ie an Analytical Run to be traced back through the sample preparation workflow to the original biological tissue collection There are fields to describe the genotype of the harvested organism as well as the growth environment and experimental treatment applied to that organism In addition there are fields to describe sample preparation and analytical protocols New fields may be added to the file without interfering with its use by MetabolomeExpress which only uses certain core fields for data processing Probably the best way to understand the format is to read the section in this manual on interpreting validation templates You could also download and examine the METADATA txt files from public experiments in the repository using the Database Navigation panel on the left of the MetabolomeExpress interface 1 6 2 Raw GC MS data NetCDF and the MetabolomeExpress eXtracted Ion Chromatogram XIC format The primary raw GC MS data format used by MetabolomeExpress Is the open standard NetCDF AIA ANDI format CDF These may be exported from most instrument manufacturer s data processing software We have successfully tested MetabolomeExpress with GC Quad
100. mparisons Please select the G x E comparisons of interest At Ler Cell Suspension Cell suspension x None At Ler Cell Suspension Cell suspension x None Experimental Design At Ler Cell Suspension Cell suspension x AntimycinA 1h At Ler Cell Suspension Cell suspension x Methanol 1h Experimental Design At Ler Cell Suspension Cell suspension x Methanol 1h At Ler Cell Suspension Cell suspension x Water 1h Experimental Design At Ler Cell Suspension Cell suspension x Water 1h At Ler Cell Suspension Cell suspension x Water 1h Experimental Design At Ler Cell Suspension Cell suspension x AntimycinA 3h At Ler Cell Suspension Cell suspension x Methanol 3h Experimental Design At Ler Cell Suspension Cell suspension x Methanol 3h At Ler Cell Suspension Cell suspension x Water 3h Experimental Design At Ler Cell Suspension Cell suspension x Water 3h At Ler Cell Suspension Cell suspension x Water 3h Experimental Design At Ler Cell Suspension Cell suspension x AntimycinA 6h At Ler Cell Suspension Cell suspension x Methanol 6h Experimental Design At Ler Cell Suspension Cell suspension x Methanol 6h At Ler Cell Suspension Cell suspension x Water 6h Experimental Design x Submig E m Ca ad c CH 0 wo t a T D 2 o o 3 O c lc ue D is W o 3 KN Ka DI ceo p c 2 reus e So D c O
101. n M Carrel ad Delon E Ti ti her Cell Seager eg Complex 1 le Treinen el NES eux M wi d gen Erba ol IE LS Here EL Hotecone Lh A moomi TERS aod mefuxme SC redeo Ed LUPUS wee cut Adee J Carrel West corca k gcten wah Be Cell Zero e Gage x 275 IDAST n11575 8 ME us enger Loi TADATA Dit Med MODUS math rotenone D imis i hari E ru mc ene pu dne SS 2 Dow Doe Most of the information here is self explanatory but it is worth pointing out a few features of this table Firstly you can sort the table according to any column by clicking on its header Secondly you can load any of the retrieved experiments into the Experiment Explorer module by clicking on the little green flask icon A next to the experiment name Thirdly double clicking on any colour coded fold change value will load the underlying raw GC MS signal regions into the Raw Data Viewer of the Experiment Explorer so that you can manually verify the automatic signal processing results The screenshot below shows the result of double clicking the top result in the result set shown above The chromatographic overlay shows that the m z 288 quantifier ion of 2 Ketoglutaric acid methoxime 2TMS is clearly more intense in the 30 mM H2O treated samples compared to the mock treated samples and the visually determined intensity ratio agrees quite well with the automatically determined fold change of 3 92 listed in the database ZC ren J Experiment Explorer Arabidop
102. n the latitude longitude format where latitude and longitude are each represented in the format degrees minutes seconds where degrees may be a ve or ve integer between 90 S or W and 90 N or E minutes is a ve integer between 0 and 60 and seconds may be a ve float between 0 and 60 For example New York City is at around 40 42 51 36 74 0 21 49 Configuration Parameters None Example Code v Llatlong Example Valid Value 40 42 51 36 74 0 21 49 Email e Data Type Code e Validation Description A valid email address Configuration Parameters None Example Code v e Example Valid Value joe bloggs foo com Example Invalid Value joe bloggs at foo com Name name Data Type Code name Validation Description A variable number of words with at least 2 words Only letters are allowed not numbers Configuration Parameters None Example Code v name Example Valid Values John Smith John R Smith The Artist Formerly Known as Prince Example Invalid Values John Prince John Smith 1 URL url Data Type Code url Validation Description A valid URL Configuration Parameters None Example Code v ur1 Example Valid Value http www ncbi nlm nih gov pubmed Example Invalid Value nttp www ncbi nlm nih gov pubmed Variable Words vw Data Type Code vw Validation Description A variable number of words with at least the minimum number of words Numbers are all
103. ne as long as the ID for each entry in the library is unique NOTE One important issue with naming however is that certain naming styles allow MetabolomeExpress to derive the name of the underivatised metabolite from the derivative name Name recognition is case insensitive The general syntax for naming is Common Name of Metabolite space Derivative Information where Common Name of Metabolite any commonly used name for the metabolite MetabolomeExpress has a database of over 100 000 different metabolite synonyms so if you use a common name it will probably be recognised You can check whether your metabolite entries are being recognised later using the MetabolomeExpress MSRI Library Manager and Derivative Information any combination of the following general terms in any order where X any integer or is omitted altogether methoxime methoxyamine MXX X TMS XTMS X TMS XTMS X TBS XTBS X TBS XTBS Peak EZ Peak X PeakX Peak X major minor BP BP BP derivative unknown derivative unknown derivative The table below shows a few examples alpha ketoglutarate 2TMS methoxime alpha ketoglutarate D Glucose MX1 5 TMS D Glucose Alanine 2TMS Alanine Glucose methoxime BTMS EZ Peak 1 Glucose RI The Kovats retention index of the analyte Please use only true Kovats RI values Quantifier lons One or more nominal mass quantifier ions to use separated by com
104. nerated wing the sample nferrnaben that you entered on the prewous screen These may be downleaded using the inks provided These matrices were buit m a total af 4 seconds Total number of missing peaks fled by raw data assisted messing value replacement peux TRE te ee Ferr Ce baies ol ei rari beet i ele sg nier ora Fer vir abor PETE PAE lute E meti mar ultri ul mma ha ag ud Re mor usus Has MM muri en abba Asp int cci dais ask Esch beta nadie daba Kin g ieren Serbien kache Leg skagen T june id i eit agp cad vigo ie gege a Tap THAE peed Be fa aAA TRTE Loved gel Ball Bae pha areas i see of Mea iin ills ol Hop rali Firei kass ones das the rules eal e bee Bal r idein ali E dein Flea ie Kan pel Tharehiea ali caeci dcn i baton d and E Tha koram a ukha ar Vaal vera gs eral interitus Ara dx pit ve roc tha ipa nd pehin A rede Pr Tha a just lica i TMA TR Bret dee riii S hiing a piik ish ini dich cf f coni celi SF Pha coat Ead Da paak von Ba r tacbon Biia and Pha bine nda date Ca bbird opami DI degree ATI sacan bey fot alee UL Tha etri Forreat cae hob Omg be uiae For Further broomino wing MatabcirwEsrewis but ii ciadul For dere peii robr rege Ber pea datectess doch id pected RE Eh aant iT Samus is MERI aequa Wegen duh 2203 Lee lancom MATRIS sre TREX Tis a Han inemanl estrzsle lem See Ripe ifs fk ge Reder de Herojai D z rk lcm Han ET brzesgl ba eosin A army amie kclesrzatere besche rier asf olala E
105. nough background let s load the data 3 2 Loading a dataset 3 2 1 Loading a dataset from the Navigation panel On the left side of the MetabolomeExpress interface you should see a panel called Navigation with a directory tree in it This is where you browse through and load available datasets First expand the root node so you can see the experiment folders indicated by little green flasks including the folder corresponding to the example dataset Then load the example dataset by left or right clicking on it to access the context menu and then left clicking on Load Folder Contents Navigation L Database Navigation j CH PlantEnergyBiology Internal 355 PlantEnergyBiology Public d a Bb sranidopsis AOX1a KO Moderate High Light and Drought Stress d A Arabidopsis Cells Rotenane Timecourse d C ZA Arabidopsis GST RNAI SA and H202 Treatments d C A arabidopsis SAL1 Mutants alx8 and fry1 1 d C lusibus mtETC CplxI Mutant ndusf4 Cmplmnt day and night E Se mtETC Cplx Mutants ndusf4 and ndufat night imecaourse m Load Ce Contents import sample Information From MIMIMET TXT P b Rice Germination Timecourse Air vs N2 d C a Rice Germination Timecourse Air M2 Gas Switch i J ia Rice Germination Timecourse M2 Air Gas Switch Wait a few moments while the information about the experiment is retrieved from the server and loaded into the interface When everything is ready the main tab panel sh
106. older it will be applied to all experiments in your repository except where locally overridden by an analyte annotations table in an experiment folder A file is recognised as an analyte annotations table when its file name starts with the string analyte annotations table If more than one analyte annotations table is detected in a folder then the one with the highest alohanumeric ranking is used eg a file named analyte anotations table 2010 txt would be used in the presence of another file named analyte anotations table 2009 txt 1 6 Supported Data Formats including example files 1 6 1 The main MetabolomeExpress metadata exchange format METADATA TXT For public dissemination it is essential that metabolomics datasets include sufficient metadata to allow other researchers to understand the biological and technical origins of the data in enough detail to be able to reproduce essentially the same results The Metabolomics Standards Initiative MSI has outlined minimal reporting standards and guidelines for metabolomics metadata reporting and these guided the design of a simple metadata exchange format for use in MetabolomeExpress The MetabolomeExpress metadata exchange format is tab delimited and designed to be readable by both humans and computers whilst retaining the flexibility and extensibility required in the ever changing world of data reporting standards The file is divided into seven main subsections as indicated by the figu
107. on start time of the second peak However in most cases there is no second peak rising out of the down slope of the current peak and the algorithm continues recording the peak until the absolute value of the negative slope falls once again below the critical slope threshold This event tells the algorithm that the end of the peak has been reached and it stops recording the peak and keeps moving along waiting for the slope to rise above the critical slope threshold again This process is continued all the way to the end of the EIC until all signals resembling peaks have been recorded In the second phase each signal section recorded in the first phase is first examined in a number of ways The retention times scan numbers and intensities of the signals at the start and end points of the recording are recorded he retention time peak apex retention time and intensity of the scan having the maximum signal intensity the peak height are recorded e The sum of all the recorded intensities the peak area is calculated e The proportion of total signal lying above the lowest integration point the integration point with the lowest signal is calculated the peak purity factor The algorithm then compares the values calculated as described above with the user specified thresholds and if the peak meets all of the criteria then it will be added to the PEAKLIST file along with all of its recorded characteristics If it fails to meet any one of
108. orithm then steps forward through the EIC scan by scan calculating the slope at each point until it encounters a slope value that exceeds the critical slope threshold specified by the user When this happens the algorithm is alerted to the fact that it could be running into rising section at the start of a chromatographic peak and starts recording retention time and intensity information for later integration until it encounters slope events that indicate that the end of the peak or the start of a new peak has been reached If the algorithm is in the rise of a peak it waits for the slope to become negative indicating that the apex of the peak has been reached and the algorithm is now entering the falling part of the peak As the algorithm moves down the falling part of the peak it keeps recording the peak until the absolute slope value exceeds the critical threshold again When this happens the algorithm checks whether the slope is negative which tells the algorithm that it is well past the top of the peak where small subcritical but transient negative slopes might be encountered and should start looking for the end of the peak or positive which tells the algorithm that it has encountered the rising part of a new peak that starts part of the way down the first peak If the rising part of a second peak is detected in the down slope of a current peak the intersection point is given as the integration end time of the first peak and integrati
109. ould switch to the MetaData Viewer and all the standards compliant metadata associated with the experiment should be visible something like this A Mozilla Firefox S E UC A Taperumaat Fxplorsr i rampla D aperunment Araheiwmis Cell Suspension intimin A Timer rum lutzhuma f mm o i A Lao Platania Viewer Bae Dei Yess Dua Jee and Dee Ddacioon BAL Loewe HERAS BSS aed Deti Lee jtetansis piini ADMINISTRATION a Experiment Name Timecouree Treatment of Arabidopsis Cell Suspension Cultures with Antimycin A Methanol or Wiater Project Name Mitochondrial Ceyshunctiony Satan hate Aram lames Caro EMT rye carroad gstudent uwa edu au TU Adam James Carrol oly ele aha carroad 1 etudent uva edu gu 4 ean Chemical inhibition of the mitochondrial eectron transport chain wil lead to widespread changes in the metaboome 120 mi Arabidopsis cell suspension cultures ecotype Landsberg erecta were treated with ether Antimycin A an inhibitor of mrETC Complex TIT dissolve in methanol methenol a carrier solvent Bref Descrption of control or water a negative control Five biological replicate cel culture flasks were treated per treatment Alquots of the cell suspension cultures were collected by pipetting and bref vaccuum fitraton Experiment dia remove and collect culture mediurm EN prior to treatment time zero and then at 1 3 6 12 16 and 24 hours alter treatment Changes in thier Ed ee celis were thin 7 studied b
110. owed but not counted as words Configuration Parameters An integer specifying the minimum number of words Example Code v vw 3 Example Valid Values Blue is beautiful Blue is really beautiful Example Invalid Values Blue Blue is John Smith 1 Variable Words Row Release vw rr Data Type Code vw rr Validation Description A variable number of words with at least the minimum number of words with a particular value releasing any further metadata requirements for the fields to the right of this field in the row Numbers are allowed but not counted as words Configuration Parameters x release_word where x An integer specifying the minimum number of words and release_word a word that if found it this field releases any requirements for metadata in fields to the right of this field in the row Example Code v vw rr 1 WT Example Valid Values WT Blue is really beautiful Example Invalid Values 1 Variable Words or Numbers vwn Data Type Code vwn Validation Description A variable number of words with at least the minimum number of words Numbers are allowed and counted as words Configuration Parameters An integer specifying the minimum number of words Example Code v vwn 2 Example Valid Values Cabinet 1 Cabinet A Cabinet A in room 106 1224 9999 Example Invalid Values 1234 Cabinet_1 Cabinet_A Species sp Data Type Code sp Validation
111. p 1 Step 3 Based on results of Step 2 does the temporarily merged MST contain at least the minimum number of expected ratio qualifier ions AND have an average ion intensity deviation below the specified threshold IF YES Add match details to tab delimited MATCHREPORT file using the integrated peak area of the quantifier ion as the reported signal intensity for the matched analyte IF NO Discard MST and continue 5 2 Interacting with MSRI library matching results When an authorised user submits one or more PEAKLIST files for library matching each PEAKLIST file gets searched for peaks matching library entries and positive matches across the entire set of PEAKLISTs are reported in a tab delimited MATCHREPORT file which appears in the experimental folder remember to reload the experiment or hit the Refresh button on the Statistics and Data Exploration control panel in order to see the report in the relevant control panels If the Display Results option is selected you will be presented with a screen like this for this example we have selected the PEAKLIST file called 030407 ANTIA 0H MEOH 0H 1 PEAKLIST which corresponds to the first biological replicate cell culture flask sampled just prior to being treated with Antimycin A the time zero time point To try the example set the library matching control panel up as shown below and hit the GO button Library Matching Control Panel Please select your PEAKLIST file for l
112. p 1 Wildtype Disease drug week 1 urine I 100 3 151009 Wildtype Disease Drug Week 1 Rep 2 Wildtype Disease drug week 1 urine 1 100 4 151009 Wildtype Disease Drug Week 1_Rep 3 Wildtype Disease drug week 1 urine 1 100 5 151009 Wildtype Disease Placebo Week 1 Rep1 Wildtype Disease placebo week 1 urine 1 100 6 151009 Wildtype Disease Placebo Week 1 Rep2 Wildtype Disease placebo week 1 urine 1 100 7 151009 Wildtype Disease Placebo Week 1 Rep3 Wildtype Disease placebo week 1 urine 1 100 8 151009 Wildtype Healthy Drug Week 1 Rep 1 Wildtype Healthy drug week 1 urine 1 100 9 151009 Wildtype Healthy Drug Week 1 Rep 2 Wildtype Healthy drug week 1 urine 1 100 10 151009 Wildtype Healthy Drug Week 1 Rep 3 Wildtype Healthy drug week 1 urine 1 100 11 151009 Wildtype Healthy Placebo Week 1 Ren Wildtype Healthy placebo week 1 urine 1 100 12 151009 Wildtype Healthy Placebo Week 1 Rep2 Wildtype Healthy placebo week 1 urine 1 100 13 151009 Wildtype Healthy Placebo Week 1 Rep3 Wildtype Healthy placebo week 1 urine 1 100 14 151009 Wildtype Disease Drug Week 4 Rep 1 Wildtype Disease drug week 2 urine 4 100 15 151009 Wildtype Disease Drug Week 4 Rep 2 Wildtype Disease drug week 2 urine 4 100 16 151009 Wildtype Disease Drug Week 4 Rep 3 Wildtype Disease drug week 2 urine d 100 17 151009 Wildtype Disease Placebo Week 4 Rep1 Wildtype Disease placebo week 2 urine 4 100 18 151009 Wildtype Disease Placebo Week 4 Rep2 Wildty
113. panel 31 2 2 Finding experiments of interest using Hesponset nder SS 2 3 Comparing metabolite response patterns across multiple publications using WEE TAG cls TR 33 2 4 Identifying phenocopies using PhenoMeter in development 36 3 Processing and analysis of experimental GC MS datasets with Experiment ic eT Q 38 3 1 Example dataset Timecourse metabolomic analysis of plant cells treated with NU ele e TT 38 3 2 Loading d SIL aceeseoisesdea copo ee rne soiNE uU pUSU Tao FED Stu E UT EE EEE tO EU Scu IN SUUS EE 38 3 2 1 Loading a dataset from the Navigation panel 38 3 3 Using the Raw Data Viewer 39 3 4 The raw data viewer control panel nnannnnnnennenennnsnnnennrnnsnnsnnrnnnnnesnrnnererenne gt 40 OTN AN RY NETTES 41 3 6 Viewing mass spectral scans cccccscccsseccececeeeceeccececseeceueceeeecseeseueeseeesaees 42 3 Finding differentially expressed peaks using the chromatographic statistical tool 42 4 Data import and peak detection registered users only 43 4 1 A Guide to the MetabolomeExpress PeakFinder Algorithm 44 42 The Peak Detection Control Palel ucc otro eerte to etre e aee ta Ernte ae erar teres 45 SME on aE n o RTT mm 46 5 1 How to conduct an MSRI library matching process 46 Mass Spectral Tag MST reconstruc
114. pe Disease placebo week 2 urine 4 100 19 151009 Wildtype Disease Placebo Week 4 Rep3 Wildtype Disease placebo week 2 urine 4 100 20 151009 Wildtype Healthy Drug Week A Rep 1 Wildtype Healthy drug week 2 urine 4 100 21 151009 Wildtype Healthy Drug Week 4 Rep 2 Wildtype Healthy drug week 2 urine 4 100 22 3151009 Wildtype Healthy Drug Week 4 Rep 3 Wildtype Healthy drug week 2 urine 4 100 23 151009 Wildtype Healthy Placebo Week 4 Rep1 Wildtype Healthy placebo week 2 urine 4 100 24 151009 Wildtype Healthy Placebo Week 4 Rep2 Wildtype Healthy placebo week 2 urine 4 100 25 151009 Wildtype Healthy Placebo Week 4 Rep3 Wildtype Healthy placebo week 2 urine 4 100 Sometimes the disease state is more to do with genotype For example if you were doing an experiment to compare the metabolomics responses of a normal mammalian cell line and some mutated cancer cell line to a treatment like say hypoxia your file might look something like this WI A B C D E Organ or Sample Sample ID Genotype Treatment Biomaterial Timepoint Mass or i Type Volume 2 151009 wildtype hypoxia Rep1 wildtype cell line Hypoxia cultured cell 1 100 3 151009 wildtype hypoxia Rep2 wildtype cell line Hypoxia cultured cell 1 100 4 151009 wildtype hypoxia Rep 3 wildtype cell line Hypoxia cultured cell 1 100 5 151009 wildtype hypoxia Rep A wildtype cell line Hypoxia cultured cell 1 100 6 151009 wildtype hypoxia Rep5 wildtype cell line
115. ration start and end points and peak purity factor defined as the proportion of the total integrated signal that lies above the lowest integration point The algorithm works in two phases In the first phase the algorithm moves from the start of the EIC to the end recording sections of the signal that resemble chromatographic peaks In the second phase the algorithm checks each of the recorded sections to see if it meets the user specified criteria for being a real chromatographic peak min peak area min peak width min peak height and min peak purity factor These user specified parameters should be optimised whenever data from a new instrument type or brand is processed Once peaks have been detected you can review the peak detection results by visualising the raw data in the raw data viewer with Display Peak Detection Results turned on The first phase begins by starting at the beginning of the EIC taking a 3 point moving average of the signal intensity centered around the second scan point ie the average of the signal intensities at the first second and third scan points taking a three point moving average of the signal intensity centered around the second scan point the average of the signal intensities at the second third and fourth scan points and subtracting the first average from the second average This value will be referred to as the slope of the signal in this case between the second and third scan points The alg
116. re below EXPERIMENTAL METADATA ADMINISTRATION METADATA BlIOSOURCE METADATA CHEMICAL ANALYSIS METADATA GENOTYPES BOS AMPLE PROCESSING PROTOCOLS BIOS AMPLE EXTRACTION PROTOCOLS ANALYTICAL SAMPLE PREPARATION PROTOCOLS METABOLITE IDENTIFICATION METADATA METABOLITE IDENTIFICATION PARAMETERS QUALITY CONTROL METADATA QUALITY CONTROL PARAMETERS DATA PROCESSING METADATA DATA PROCESSING PARAMETERS STATISTICS METADATA STATISTICS PARAMETERS The Administration Biosource and Chemical Analysis metadata sections are the only ones essential for MetabolomeExpress processing Their structures and content are shown below ADMINISTRATION METADATA ADMINISTRATION Experiment Name Project Name Biological Experimentalist Name Biological Experimentalist Email Metabolome Analyst Name Metabolome Analyst Email Experimental Hypothesis Brief Description of Experiment Literature Reference Journal Publication Date PubMed ID BIOSOURCE METADATA GENOTYPES Genotype 1 Genotype ID Species name Ecotype Background Ecotype Background Catalog Marne Ecabype Background Catalog URL Transgenic Mutant Mame or WT Transgenic Mutant Name Catalog Mame Transgenic Mutant Nam Catalog UAL Relevant Genes Description Genotype 2 ENVIRONMENTS Environment 1 Environment IL Location Growth Protecel Description Plot Design Light Period Hum
117. re the metabolite response patterns of 4 experimental class comparisons one for rice seedling anoxia at a 48 h timepoint and three for poplar root flooding at 5 24 and 168 h timepoints MetaAnalyser Control Panel a Please select metabolite responses from the tree Options Filter metabolites of unknown structure _ 4 Narsai R Howell KA Carroll A lvanova A Millar AH Whelan J 2008 Defining core metabolic and transcriptomic 1 a Or E Rice germination timecourse germination in air ws germination in N2 a 3 Rice germination timecourse Air N2 gas switch E Rice WT seedling Ar N2 Switch 24h Rice WT seedling Air_24h E E E Rice WT seedling Ar N2 Switch 25h Rice WT seedling Air_25h E Rice WT seedling Air N2 Switch 27h Rice WT seedling Air_27h Iz E Rice WT seedling Ar N2 Switch 30h Rice WT seedling Air_30h Rice WT seedling Air_N2_Switch_48h Rice WT seedling Air 48h b gl Rice germination timecourse N2 Air gas switch 3 t C Sappl PG Carroll AJ Clifton R Lister R Whelan J Harvey Millar A Singh KB 2009 The Arabidopsis glutathione tr t Howell KA Narsai R Carroll A lvanova A Lohse M Usadel B Millar AH Whelan J 2009 Mapping metabolic and 4 Kreuzwieser J Hauberg J Howell KA Carroll A Rennenberg H Millar AH Whelan J 2009 Differential response 4 V Timecourse Flooding of Poplar roots Populus x canescens Root Flooding 5h Populus x canesc
118. rms for fungal anatomy Pans pune http www yeastgenome org fungi fungal anatomy ontology iidescripti on Obtained from http www obofoundry org cgi bin detail cgi id fungal anatomy Human Developmental Anatomy Ontology Abstract Version terms Anatomy human anatomy Obtained from http www obofoundry or i bi il cgi i dev anat abstract dmelanogaster genes E 6 1 on 2010 04 02 Systematic names of yeast Saccharomyces cerevisiae genes eg YDLO78C as per the Saccharomyces Genome Database SCH http downloads yeastgenome org chromosomal feature SGD features ab l Gene Names for genes in the Escherichia coli K 12 genome eg modB as ecoli_genes per the EcoGene database http www ecogene org Mouse Adult Gross Anatomy Ontology http www informatics jax org searches AMA form shtml terms mouse anatomy Obtained from http www obofoundry org cgi bin detail cgi id adult_mouse_anatomy FlyBase Drosophila Gross Anatomy Ontology terms for fly anatomy fly_anatomy Obtained from http www obofoundry org cgi bin detail cgi id fly anatom Plant Ontology Consortium www plantontology org terms for plant plant_development growth and development Obtained from http www obofoundry org cgi bin detail cgi id2po temporal EES ENER FlyBase Drosophila Development Ontology terms for fly growth and fly development development Obtained from http www obofoundry org cgi bin detail cgi id fly development Human
119. rol Panel Match Report to Data Matrix Data Matrix Renormalisation Matrix Explorer Comparative Statistics PCA HCA Correlation Analysis ResponseFinder ReponseFinder MySQL DB Stats Import Please select a peak area matrix for renormalisation GCMS_ID Batch Repor_11 August 2008_5 14 47 mztMATRIX Normalise each sample to the mean of its control values Wait a few moments and you will be presented with the renormalised matrix displayed as an interactive heatmap like this 4 2 dei 17 hat focalhost Metaboiomeexpress si ki K al a ty H metabolome x p r e S S THE ONLINE TOOLKIT FOR HIGH THROL LE TET ava Data Verve Data Irgat ad Bes Don bentur HERI Lirary Matchen Slalintics and Daka Fxplsralizon neak Chart part Normalisation Overview z Your matrix i amp called GCMS_ID Batch Report 11 August 2008 5 14 4713 August 2008 5 56 9 mzrtMATRIX RENORM You can download it using this link 030407 030407 030407 030407 030407 E AMTIA ANTIA ANTIA ANTIA AHTIAUT 20407 030407 030407 030407 030407 03 METH METH METH METH METH W J HMEO OHMEO OHMEO OHMEO OHMEO Z NOLO NOLO ANOLO ANOLO NOLO GO HOH1 PHOHZ PHOHS PHOH4 PHOHS P ve PEANZ PEAHJ3 PEAHA PEAHS PEAEA Bereet KLIST KLIST KLIST KLIST 51 5 3 2 SLT A E rem grind re ney nen ein M tan Aran S een Suspension Suspension Suspension Suspension Suspension Suspension Suspension Suspension Sus Call Coll Cell Cell Call Noni H ig WEE Pd iba f TE
120. rtimycina 24h Al Ler Cell Suspension Ler Cell Suspension Ler Cell Suspension Ler Cell Suspension Ler Cel Suspension Ler Cel Suspension cultured cell cultured cel cultured cell cultured cell cultured cell cultured cell Methanol 1h Methanci Jh Methanci 6h Methanol 12h Methanol 16h Methanol 24h Average TN Signal Signal Signal Signal Signal Signal a Retention Quantifier y Analyte Signal Annotation Retention Intensity pal Intensity pal Intensity paal Intensity p val Intensity pa Intensity Index mz ime min Kovats Ratio Ratio Ratio Ratio Ratio Ratio Amino Acid L Glutamine ITAS 10190 RI1775 6 MZ156 27536 17753 1 0 22 T 0 12 0 05 0 13 ino Acid L Asparagine 3TMS 10164 RI16735 MZ188 0 23 0 14 0 22 7 zi ino Acid L Asparagine 3TMS 10163 RI6732 MZ188 168 16844 i T 21 222 0 23 0 14 0 23 T Genotype x Eneronment Comparison ino Acid L Glutamic acid 3TMS ID146 Ri16207 Mz245 24644 16206 246 0 17 24 0 33 0 28 0 15 Amino Acid 4 Hydroxyproline 3TMS ID114 RI15195 MZ244 244 0 89 0 47 0 38 1 38 ino Acid L Ornithane 4TMS 10197 RI18134 Mz174 28224 1813 174 j4 d 0 69 he 134 0 53 0 57 0 77 Amino Acid Pyrogiutamac acd 2TMS Di R11521 2 M7156 156 0 62 0 53 0 7 5 t 17 ino Acid Pyroglutam c aGd 2TM3 _10116_RI1521 2_MZ156 22786 156 27017 0 64 0 53 0 7 Amino Acid L Alanine 2TMS ID15 RIT0014 MZ116 1204 14 1 26 2 2 52 L Alanine 2
121. rupole MS NetCDF files exported from Agilent s ChemStation software and GC TOF MS files exported from LECO s ChromaTOF software Slight differences do exist between the structures of CDF files exported from different types of instruments Please contact us if you have any problems with your files and we will fix them Before you can work with your raw data in MetabolomeExpress you must import your CDF files through the generation for each CDF file a corresponding file in the custom MetabolomeExpress eXtracted lon Chromatogram XIC binary format Unlike NetCDF files which are indexed by scan number XIC files are indexed by m z channel Therefore MetabolomeExpress rapidly retrieves scans of interest from NetCDF files and rapidly retrieves chromatograms of interest from XIC files Details on how to import raw data files are given in the section in this manual called Data import and peak detection registered users only 1 6 3 Peak list tables PEAKLIST A peak list table file extension PEAKLIST contains information about all the extracted ion chromatogram EIC peaks the signal peaks in each nominal mass m z channel in a given GC MS chromatogram Peak lists are simply tab delimited tables with the following columns m z the integer m z value of the EIC peak Apex Time The retention time of the peak apex in minutes Integration Start Time The retention time of the start of the peak ie the point at which the rising signal first
122. s Standards Initiative MSI and some recommendations have already been made in the form of roadmaps for standard reporting in key metabolomics fields http msi workgroups sourceforge net To ensure that datasets contained in the main metabolite response database are annotated in a systematic and standardised manner that meets the recommendations of the MSI METADATA TXT files must be checked against an appropriate validation template before the associated datasets may be imported into the database These validation templates define e The names of metadata fields e Which fields are mandatory and which fields are optional e The range of allowed values that a field may be assigned e Written instructions for appropriate completion of each field 7 2 Interpretation of MetabolomeExpress Metadata Template Files Metadata template files have essentially the same structure as the MetabolomeExpress METADATA TXT metadata format except special codes appear where metadata values would normally be These codes define the acceptable values that may be given for their respective fields For example the table below shows the ADMINISTRATION section of one metadata validation template MetabolomeExpress Experimental Metadata File Mammalian Metabolomics v1 0 ADMINISTRATION METADATA ADMINISTRATION OO Field Experiment Name v vwn 1 Give the name of the i experiment here v vwn 1 Give the name of th
123. sample Different treatment durations or treatment doses are considered different treatments and must be given different IDs Make sure that all samples of the same treatment have exactly the same entry here Organ or Biomaterial Type This is the standard name of the organ tissue biofluid type that was used to prepare the sample Check reporting standards in your biological field for the appropriate controlled vocabulary or ontology to use here Make sure that all samples of the same organ or biomaterial type have exactly the same entry here Timepoint This is the time of harvest of the sample with respect to the beginning of the treatment period Make sure that all samples of the same timepoint have exactly the same entry here Sample Mass or Volume This is the mass or volume of the biological sample that was extracted and analysed The units do not matter at this stage They may be added to the more detailed METADATA TXT file later These numeric values will be used to normalise signal intensities later during data matrix construction We will now run through some examples to demonstrate how you can set up MINIMET TXT files for different types of experiments The screenshot below shows an example MINIMET TXI file being created in Microsoft Excel You can probably tell that the file represents an experiment where two genotypes of animal have each been fed on two different diets and their blood has been collected one day and two days
124. sierto epe eee rere s fuus cx n UD ADR da c ap aman ert Us The following correlation network files have been generated These may be downloaded using the links provided EN CHEM NNNM 7 ON NN Download Link GCMS ID Batch Report 11 August 2008 5 14 47 CYTOSCAPE CORNET 19 August 2008 4 35 21 TXT GCMS ID Batch Report 11 August 2008 5 14 47 PAJEK CORNET 19 August 2008 4 35 21 NET This format consists of a square tab delimited correlation matrix giving all analyte analyte correlation coefficients This is the same correlation matrix as described above except that the cells Correlation Matrix containing correlation coefficients have been colour mapped so that positive Colour Mapped correlations are blue negative correlations are red and the intensity of the HTML Table colour is directly proportional to the strength of the correlation It may be opened in MS Excel GCMS ID Batch Tab Delimited Report 11 August 2008 5 14 47 CORMAP 19 August 2008 4 35 21 TXT GCMS ID Batch Report 11 August 2008 5 14 47 CORMAP 19 August 2008 4 35 21 HTML You can download the various output files using the provided hyperlinks To import the Cytoscape compatible file into Cytoscape go to the File gt Import gt Network from Table Text MS Excel and set the dialog box up like this HINT be sure to unselect Space as a delimiter under Text File Import Options before trying to select columns amp Import Network and Edge Attributes from Table
125. sing rh ink GST_ GST G5T GST G5T n GST_RGST_RGST_RGST_ RGST R RNA RAAT BRAL DRAI RBA Quai Eat RHAL Rear Side NAL MAIL 32HAI 38HAI 3NAI 3G5T RN 1HM 1HH 1HH 1HHM 1HHMH LHM IMM IMH IMH IHH MH Dnn DHM UMA HMHM AI Ho Sa 3 5A 3 5A 35A 35A 3 SA 35A j5A 35A Zen 3 202 H202 H202 H202 H202 CK 210 MIN MIN OMIN OMIN OMIN 4H i 4H 3 4H 3 AH 4 4H p AAH 24H Aan AH 24H MIN 1 z PE 2 PE 32 PE APELS VEGEAKRKIPEARKLPEAKLPEAKLPEARLZIIDE 2 PE i PE PE S PEPEAKLI K ISAELIBAKLISARLISARLIS 5T ERREUR ia AKL 1 Li E IST IST IST 1ST IST 1 T E L E a0 1H e RMNGST SRAGST RUAGST NGS ANIAGST RUGS UST ANUN GST RUST ANN GST RUST NGST RIA GE seeding seeding seeding seeding ten 530mm som stm MAA a 24h sa 24h sa 24h Ee Wr 0 5 DA H EN 7 E i 1 1 KS T 1 1 1 20 30 D 3 E ri 2 3 4 5 1 d 3 i d i 4 5 1 Pyrictinia 3 Frmathawiilazeg Ms RILOOU 4 DDG RIlODO 4 MII52 2 2 Phenyi A 2H quinazolinane 3 ybneethyiithio 4 3H quinazolinone MSG RI1022 8 105 RI MII Ghroolic ack ATMS I AOS MZ147 Acetophenone MSUA RI6 7 11 DGIO RILIUSZ MILOS 11 285 11 962 12 085 m Match cae Once you have generated a matrix in this way you can either apply another metadata based normalisation or move straight on to some multivariate analysis statistics or correlation analysis 6 2 Matrix renormalisation By default all mzrtMATRIX format data matrices are already normalised to internal standard pea
126. sis GST ERAI SA and A909 Treatments 8 conem E e p negetanmn Meiri Vewer Bee Daba Viewer Data imootend Pesk Detector MSAD Library Matching Staitstcs and Date Exploration EIC Overlay RT 23714 1e 23 014 min mz 208 ZRekemre acid matheeime GOTHSA WIR ANTES MES WI SOMM HAO 74H 1 WT 30MM H202 24H 2 WI SOMM HOCH 24H 3 WT 30MM H202 24H 4 I OMNI ES D JH 5 Intensity VN WLR SGH wur ow ee LA 1 0854 de 1L 1 JAM A LA TI Y Uum ao x 006 1 dn WE fg 2 i ER ER SR 2 5 Comparing metabolite response patterns across multiple publications using MetaAnalyser The MetaAnalyser module allows you compare metabolite response patterns across different experiments and different publications MetaAnalyser assembles the metabolite response profiles from selected experiments assembles them into a data matrix and carries out a 2 way hierarchical clustering before returning the organised results in the form of an interactive DHTML heatmap and a PDF clustergram MetaAnalyser also scores metabolites according to their responsiveness ie how much variation they show across the selected dataset Using MetaAnalyser is very simple You simply select which experimental class comparisons you wish to include in the analysis specify whether to include metabolites of unknown structure and click GO The screenshot below shows the MetaAnalyser control panel set up to compa
127. sting peaks you can use the chromatographic statistical comparison tool To see an example set the Raw Data Viewer Control Panel up like this it doesn t matter which runs are selected in the bottom part and hit Display Raw Data Viewer Control Panel Chromatogram Selection Blue Chromatogram 5 Red Chromatograms O6040 METHANOL 24H 1 XIC D6040 METHANOL 24H Za D60407 METHANOL 24H 3 XIC D6040 7 ANTIA 24H MEOH 1 XIC D60407 ANTIA 24H MEOH 2 XIC D6D4D ANTIA 24H MEOH 3 XIC D6D4D 7 ANTIA 24H MEOH 4 XIC O60407 ANTIA 24H MEOH 5 XIC D6D407 METHANOL 24H 4 XIC D6D407 METHANOL 24H 5 XIC This is what the view should look like after clicking the Display button and selecting one of the runs for display of peak annotations Chromasographic View TIC Overtay RT D to 60 mn j j m og Wn I I 20088 B it it il II i vw v TU W YCW WT wee ee ee ee reg rer emm r gr I0 00 Y H 160407 ANTIA 24H MEO 107 ANTIA 24H MF OH NTIA24H MEOH UNT IA 74 H Mi t NTIA24H ME I r Intensity 0 00810 30 40 50 60 Retention Time min Notice the little red and blue markers at the top of the chromatogram These indicate scans that are statistically significantly higher in the red or blue chromatograms respectively If you zoom in on those peaks you will see that they are usually biologically responsive analytes Hint If you are only interested in really strongly responsive metabolites you can increase the minim
128. tically Significant a E If this box is checked detected peaks will be labeled with their retention timc Scans PESO If this box is checked scans that are significantly more intense in Ehe red or bluc sets Min Fold Change 2 Diae will be marked with red or blue marks respectively at the top of the chromatogram display Max p value 0 05 4 0 Set the statistical thresholds here If this box is checked detected peaks will be addi Once loaded select which run to show peak annotations fori Choose a library iii report containing peak identifications here e Hit this button te toad the selected data If you load some raw data with the Display Library Match Results option selected the names of loaded runs for which library matching results are available from the selected match report will appear here io display the peak identifications just click on the run of interest Litle green arrows should appear at the top of the display You can mouse over these to see the library match results for the peaks below Now set the Raw Data Viewer Control Panel up as shown in the image above by selecting the first chromatogram in the Blue Chromatograms selection box setting the m z to 147 and checking the Show Peak Detection Results and Display Library Match Results checkboxes Then hit the Display button Wait a moment and the selected chromatogram should appear in the viewer like this you may wish to collapse or drag th
129. tifier ion in the matched peak in arbitrary peak area units RT Apex The retention time at the quantifier ion peak apex in minutes RT Start The retention time at the integration start point of the quantifier ion peak in minutes RT End The retention time at the integration end point of the quantifier ion peak in minutes RI Apex The Kovats retention index at the apex of the peak Delta RI The retention index error of the quantifier ion peak ie Observed HI Expected RI Coverage The percentage of ion signals present in the library spectrum that are present in the extracted MST Match Details A string providing details of the library match such as score quality etc The MetabolomeExpress MSHI Library Matching alorithm provides information about the number of ion signals in the extracted MST that show the expected intensity ratio with respect to the quantifier ion within the given tolerance as well as the average deviation of all ion intensities from their expected ratios 1 6 5 Data matrices mzrtMATRIX The main data matrix format used by MetabolomeExpress has the file extension PmzrtMATRIX It is a tab delimited format arranged with metabolite signals in rows and runs samples in columns There are a number of column header rows at the top of the table containing useful metadata about runs samples and a number of row header columns containing information about metabolite signals Column header rows
130. time of each HI calibration compound in at least one representative run from your batch We use the MetabolomeExpress raw data viewer or the freely available AMDIS software to identify the peaks for a series of alkanes and enter their retention times into a template CAL file using a text editor You can easily create CAL files yourself in a spreadsheet just be sure to save as a tab delimited text file with extension cal or CAL or you could modify an example file The screenshot below shows the simple structure of the format The file has no column labels The first column is the retention time in minutes to three or more decimal places The second column is the Kovats RI These first two columns need to be filled for use with MetabolomeExpress The third fourth and fifth columns are used by AMDIS but not by MetabolomeExpress Leaving them filled is optional for use with MetabolomeExpress NM 9 G T Alkane RI 030407 Oh cal Microsoft Excel Pl X a Home Insert Page Layout Formulas Data Review View Add Ins x Calibri ni Ep General ll A 3 Insert x N Ea br orator nm epAdin e ee Ed llaa a Delete E d ves Ju A GE EIE za 28 x EK Format 7 Clipboard e Font E nues ai Number Cells Editing SI JS A B G D E F G E 1 15 461 1200 100 100 n dodecane 2 22 441 1500 100 100 n pentadecane 3 29 926 1900 100 100 n nonadecane 4 34 615 2200 100 10
131. tion and the MSRI library matching process 47 5 2 Interacting with MSRI library matching results c ccc ceecceeeeceeeeeeeeeeeeeeees 47 6 Statistics and data exploration cccceccccseccceecesseeecsececeeeeceeeseeeseueessneesseeeteeeeeas 49 6 1 Construction of data matrices from MSHI library matching reports 50 6 1 1 Some notes on normalisation and quality Control 50 6 1 2 Raw data assisted missing value replacement 51 6 1 3 How to build a data matrix using the Match Report to Data Matrix tool 51 6 2 6 3 6 4 6 5 EM renoma Salo eeren Er situs serires deese aeo EERE 54 Using the interactive matrix Cx OIG EE B5 Comparative SI SH acres cere aece secensasaroeasaissasarcmateresasascuabest nin n UNES MSIE 56 Principal Components Analysis DCAI 58 6 6 Hierarchical Cluster Analysis HCA nnne 60 6 7 Correlation network Consiruchon enne 63 6 8 Submitting a dataset to the main database of metabolite response statistics 65 APPENDIX A Interpretation of MetabolomeExpress Metadata Validation Ea o TNNT 67 Pal E E e OUD E 67 7 2 Interpretation of MetabolomeExpress Metadata Template ties 67 7 3 Validation Codes 1 00 cccceecccceecceceeceeceeceeeeeeeeseucesseeessaueessaueessaeeesseeeessnseesaeees 69 ZA Data types and associated configuration parameters 69 1 MetabolomeExpress G
132. ublic insect Metabolomics 3 Fungal_Prytopath Public Mammalian M wanes gt 3O Reviewer_A_Test_Repesitery Public jReviewar B Test Repository Public Plant Metabolomics b ganeu Aratydopsis Marans a Reviewer C Test Repository Public Omer b Rice Oryza sama O ANURS M3 Facility intemal 1 Ober 3j jFungal Phytopath Inteenal System Status Peak Detection Control Panel Please select one or more of the following data files Peak Detection Settings Refresh GST RNX MM SA 24H XIC D gebai GET RNA DIMM SA 24H 2 200 Contact Information t GST DM MM 5A 24M 350 Min Peak Arsa 1000 GST RNX MM 9A 24H ANE Licensing t GST RNA IMM SA 24H 5XIC a 500 x Finck nbtel Nes Previowt Highlight al Match eace Once you select your research field you will be asked to confirm that you want to create and new MEI ADATA TXT file and write over the old one A backup of the original file if one existed will be made under a new file name tagged with the word backup and a timestamp Overwrite existing METADATA TXT file x ch This process will overwrite any existing METADATA TXT file in ep the data directory Are you sure you would like to do this Yes l Ma l Cancel If you click OK you will be provided with a link from which to download your new METADATA file Metadata Template Download a H Your autogenerated template METADATA TXT file may be downloaded using this link NOTE If there was an existing METADATA TXT fil
133. um fold change or decrease the maximum p value settings in the control panel That s about all you need to know for the Raw Data Viewer Now let s move on 4 Dataimport and peak detection registered users only You will notice that to the right of the Haw Data Viewer tab the next tab is called Data Import and Peak Detection f you click on that tab you will see a screen like this H Mozilla Firefox DEOR File Edit View History Bookmarks Tools Help O X d wissen https www metabolome express org Es E Y g A E s L Most Visited M https www m express org M m e ta b re i re m e Currently logged in as guest xpress org login username DEE NN E Navigation ay Home Experiment Explorer Example Experiment Arabidopsis Cell Suspension Antimycin A Timecourse TD Datebase Explorer o9 Help a Registration Database Navigation MetaData Viewer Raw Data Viewer Data Import and Peak Detection MSRI Library Matching Statistics and Data Exploration 3 jPlantEnergyBiclogy Public J AOX1a KO Moderate High Light and Droug Setpet 3j J Arabidopsis Cell Suspension Rotenone Tim 3j Complex Mutant Diurnal Timecourse 180 4j jExample Experiment Arabidopsis Cell Sus 5 GST RNAi SA and H202 Treatments 3j jMetsbolomic Characterisstion of slx8 and fi 5 Poplar Flooding Timecourse dd Rice Germination Air N2 Gas Switch 4 Rice Germination N2 Air Gas Switch Peak Detection Control P
134. ur research area MetabolomeExpress provides 18 different metadata validation templates tailored to the unique metadata and ontology requirements of different research fields and different model organisms To do this go to the MetabolomeExpress interface and right click on an experiment folder in the data tree in the Database Navigation panel Then follow the Import sample Information from MINIMET TXT menu and find your research area in the list of options See the screenshot below for an example metabolome xpress org Navigation tome A Experiment Explorer Arabidopsis GST RNAI SA and H202 Treatments WR Database Semer Q Heb ZB Reostate Database Navigaton Metadata Views Raw Data Viewe Data Import and Peak Detection MSRI Lbrary Matching Statics and Data Exploratic 3 O PlantiinergyBiotogy_imternal 3 jPtntfinergyflictogy Public 3j Arabidopsis AOXTA KO Moderate High Ligty I O Arabidopsis Col Suspension Timecourse c IO Arabidopsis GST RNA SA and H202 Trest 3 O Arabidopsis Mitochondrial ETC Ceeselex M Arabideesis Mitochondrial ETC Comelec Mi Arabideesis SAL 1 Mutants saf and fry1 1 TI L ereas 4 Pon Load Folder Contents 3 Ric Import Sample Informaton From MININET TXT Choose a research area feet wa e re vs FU 1717 Bacterial Metabotomics 3 Rice Germination Timecourse AieN Gas Em remantai Metabolomics a Rice Germination Timecourse NDA Gas 3 Fungal Metabolormcs 2 ANU_RSS_MS_ Facility P
135. v LD FS 0 ge D L e o 3 CH v t c a hel D 2 o o j MN x m D n e x D OH D is 3 o 23 ic jm O o Gel j lt You will notice that a number of the possible options at the top of the list are marked with the label Experimental Design on the end These are the class comparisons that reflect the original design of the experiment as intended by the authors ie treatment vs control If you choose from these you know you are looking at results that are directly related to the original hypothesis behind the experiment Now select the Experimental Design comparisons with Antimycin A sample classes as the numerator and hit the Submit button Wait a moment and you will be presented with a heatmap statistical table like the one shown below You can sort the rows of the table according to the values in any column by clicking on the corresponding column header The image below shows the table after being sorted first by the signal intensity ratio of Antimycin A treated cells vs Methanol treated cells at 12 h and then by Chemical Class metabolome x p r e S S i ONLINE TOOLKIT FOR HIGH THROUGHPUT GC MS METABOLOMICS DATA ANALYSIS AND SHARIN MetaData Viewer Raw Data Viewer Data import and Peak Detecton MERI Lev ary Matching Statistics and Data Exploration ep Output Statistics and Data Exploration Control Panel H Statistical Report Format File Name Hyperlink Tab de
136. w The table at the top provides you with hyperlinks to download the input and output files Tab delimited PCA Input Data Matrix filtered GCMS ID Batch Report 11 August 2008 5 14 47 mzttMATRIX FILTERED 1218681255 375 Tab delimited PCA Input Data Matrix Transposed and Simplified R Importable GCMS ID Batch Report 11 August 2008 5 14 47 mzrtMATRIX FILTERED 1218681255 375 TRANSPOSED Tab delimited PCA Scores Table GCMS ID Batch Report 11 August 2008 5 14 47 mzntMATRIX FILTERED 1218681255 375 TRANSPOSED PCASCORES Tab delimited Variable Loadings Table GCMS ID Batch Report 11 August 2008 5 14 47 mzrtMATRIX FILTERED 1218681255 375 TRANSPOSED PCALOADINGS Tab delimited PC Standard Deviations Table GCMS ID Batch Report 11 August 2008 5 14 47 mzrtMATRIX FILTERED 1218681255 375 TRANSPOSED PCASD A 2D PCA Score Plot in Portable Network Graphics PNG format showing the first two principal components G x E Style Legend Ler Cell Suspension Cell suspension x AntimycinA 6h Ler Cell Suspension Cell suspension x Methanol 6h t Ka T9 WT ME TNT METHANOL 6H 2 0407_METHANOL 6H 3 0407_METHANOL 6H 1 5950407 METHANOL 6H 5 490407 ANTIA6H MEOH 6H 2 490407 ANTIAGH MEOH 6H 3 490407 ANTIAG6H MEOH 6H 1 490407 ANTIA 6H MEOH 6H 5 PC2 23 59 of Total Variance 50407 ANTIA6H MEOH 6H A PC1 43 27 of Total Variance This plot shows that PC2 contains most of the interesting biological variance related to the Antimycin A treatment
137. x Explorer tistics PCA HCA Correlation Analysis ReponseFinder MySQL DB Stats Import Please select a MATCHREPORT file LOOMS ID Batch Report 27 Julv 2009 13 27 43 MATCHREPORT pn Choose your normalisation method here Normalisation Method NOTE Peak areas will be automatically normalised to tissue mass volume given in the METADATA TXT file Normalise to single internal standard x Internal Standard Identifier String must be a unique part of the internal standard s match annotation Ribitol f normalising to an internal standard specify it here QC Minimum Internal Standard Peak Area 1000000 QC Maximum deviation of internal standard peak area from median area for same instrument batch sequence 50 S s B EA Set the minimum absolute un normalised internal standard or sum peak area here Set the maximum deviation from the median internal standard or sum peak area here E To build a matrix select your MATCHREPORT file set your normalisation and QC parameters and click the GO button You will then be given an opportunity to select which runs you would like to include in the matrix E authorisation OK File Selection Select the data files to include in analysis 030407_ANTIAOH MEOH DH 1 PEAKLIST EP 030407 ANTIAOH MEOH OH 2PEAKLIST Em 030407 ANTIA OH MEOH OH 3 PEAKLIST 030407 ANTIA OH MEOH OH 4 PEAKLIST 030407 ANTIA 0H MEOH OH 5 PEAKLIST 030407 METHANOL 0H 1 PEAKLIST 030407 METHANOL OH 2 PEAKL
138. xane Duces M532 RIT152 7 122 RIT152 77 WCS 14 34331152 2221 Ea T TER 14 E ER DE DAT Phesphoric acid ester M531 RIMTO G Putative but 44 5404 2244 PEN TE confirmed ID25 RH1TO0 6 MI E n ESAE 5 Butanaic acid 2 op eng thyleityl deriv MSEJ RI1192 8 ID27 RIT192 8 Mz147 15 2551193 2147 EA n dadecane BE Ria Mi 15 415 1500 di RES I n dodecane IDES HiT200 MES 15 413 1200 EN LB n dodecane ID29 Hit 300 WI 15413 1200 71 WEN 1 Aminocyclopropanecarboxylic acid 2TMS ID32 RITZ204 7 MI207 15 5211204 2 207 MD 1 Aminacyelapropanecarbowylic acid 2TM5 ID33 RiT205 4 MZ202 15 52411204 2 207 PEE L Valine 27M S ID35 RI DA MZ144 15 689 1271 9 144 Glycine H acetyl trimethylsilyl ester MSTT RZ a ID37 RITZ2T 58 Med 15 0651722F 3 1 4 0 29 _ Unknown P2512 ID RH2M 2 MZZM 16 178 1222 224 DIC You can download the resulting matrix using the hyperlink at the top or you can make a note of the name of the matrix hit the Refresh button on the Statistics and Data Exploration Control Panel and it will appear in the selection boxes of all the relevant tools 6 3 Using the interactive matrix explorer Before we move on to the other tools there is a feature of the matrix display that you should be aware of If you click on the file name headers at the top of the matrix you can make their background color cycle through blue red and back to white This allows you to select interesting samples out of the matrix so that if you dou
139. y GC MS analysis of complex polar metabolite extracts obtained by a hot aqueous methanol extraction method Microarray analyses of tisue samples from a parallel experiment were ako conducted on time zero 3h and 12 h samples Literature Reference Unpubisherd Transgenic Mutant Name o ransgenic Mutant Marne MT Calinog mm Ecotype j Backgoumd Catay URL At Ler Gell Weg hittpz Pero h nim nih gov DESE Landsberg erecta NCBI Taxonomy Database a r WT ENVIRONMENTS Eee EE EES SEH DEEN SE EEN ee d Ue Done l EEJ 3 3 Using the Raw Data Viewer Now if you click on the Raw Data Viewer tab metabpoiome xpress org amp Home A Experiment Explorer Example Experiment Arabidopsis Cell Suspension Antimycin A Timecourse dn Database Bale w Help 2 Registration MetaData Viewer Raw Data d Data Import and Peak Detection MSRI Library Matching Statisbcs and Data Exploration MetaData Display This is where you interact with raw GC MS data files once you ve loaded an experiment f ADMINISTRATIO You will see a screen something like this k Home Experiment Explorer Example Experiment Arabidopsis Cell Suspension Antimycin A Timecourse Database Explorer Help z Registration tit MetaData Viewer Raw Data Viewer Data Import and Peak Detection MSRI Library Matching Statistics and Exploration Chromatographic View Raw Data Viewer Control Panel A

file - BioMed Central

Contents

Download Pdf Manuals

Related Search

Related Contents