Empfohlen
Empfohlen
Weitere ähnliche Inhalte
Was ist angesagt?
Was ist angesagt? (20)
Ähnlich wie IRSAE aquatic ecology 28 June 2018 metabolomics
Ähnlich wie IRSAE aquatic ecology 28 June 2018 metabolomics (20)
Mehr von Panagiotis Arapitsas
Mehr von Panagiotis Arapitsas (14)
Kürzlich hochgeladen
Kürzlich hochgeladen (20)
IRSAE aquatic ecology 28 June 2018 metabolomics
- 2. The allegory of the Cave
- 3. Metabolomics: www.scopus.com statistics ( TITLE-ABS-KEY ( aquatic AND ecology ) ) 13200 documents ( TITLE-ABS-KEY ( aquatic AND ecology ) ) AND ( metabolomics ) 76 documents 23 reviews
- 4. Metabolomics: definitions What can the cell potentially do? What is currently being turned on? What enzymes are currently active? What is being produced/consumed? System biology
- 5. Metabolomics: definitions metabolomics is the "systematic study of the unique chemical fingerprints that specific cellular processes leave behind", the study of their small-molecule metabolite profiles (Daviss 2005) metabolomics is a newly emerging field of "omics" research concerned with the comprehensive characterization of the small molecule metabolites in biological systems. (Metabolomics Society) metabolomics is the comprehensive and holistic study of the metabolome the complete set of small-molecule metabolites to be found within a biological sample metabonome: the complete set of metabologically regulated elements in cells
- 6. Metabolomics: definitions Organic compounds of small molecular weight (less than 2000 Dalton) Sugars, amino acids, ammines, vitamins, lipids, organic acids, phenolic acids, flavonoids, stilbenes, sugar- alcohols, carotenoids, terpenoids, hydrocarbons, etc polymeric: saccharides, lignin, peptides, tannins, DNA, proteins, etc metabolomics aims to measure ALL metabolites with ONE analysis In metabolomics by definition the metabolites of interest (analytes) are NOT pre-defined Please DON’T use the term ‘targeted metabolomics’ (use the term: ‘analysis of metabolites’)
- 7. From targeted to untargeted
- 11. Targeted versus Untargeted Prefer targeted methods when: You focus in a targeted group or single metabolite (you know what to measure) You want/need the absolute concentration (except NMR) You don’t know very good your instrument (analytical chemistry skills) You are not familiar with compound annotation You have a poor knowledge of your sample metabolome You don’t have plenty of time for data analysis / you want fast results You want to work alone (biology, organic chemistry, biochemistry, analytical chemistry, bioinformatics, chemo-metrics)
- 12. Metabolomics: metabolite What is a metabolite? Product and intermediate of the metabolism – small organic molecules
- 13. Metabolomics: from where metabolites are coming from What can the cell potentially do? What is currently being turned on? What enzymes are currently active? What is being produced/consumed? System biology Are all metabolites product of biochemical process?
- 15. Metabolomic: what is and what is not How big is the metabolome? Sigma-Aldrich has ~55K commercial available chemicals Kegg contains 18K metabolites HMDB is based in ~42K metabolites Plant metabolome is estimated to cover 200K metabolites PubChem ID contains more than 10M entries CAS contains over 90M unique organic and inorganic chemicals ChemSpider contains over 57M compounds from 530 data sources 2017
- 16. Plant metabolome is estimated to cover 200 000 metabolites #ofmetabolites5-21% ethanol g/L mg/L µg/L ng/L pg/L fg/L Metabolomics: what is and what is not How big is the metabolome?
- 17. #ofmetabolites5-21% ethanol g/L mg/L µg/L ng/L pg/L fg/L Plant metabolome is estimated to cover 200 000 metabolites Metabolomics: what is and what is not How big is the metabolome?
- 18. Lisec et al. Anal Chem 2016 Currently, the best dynamic range of modern MS is 106 approximately and is significantly lower than the estimated concentration range of cellular metabolites as 1012 or more (Lei et al JBC 2011) Metabolomics: problems
- 19. Metabolomics: facts Holistic approach: complementary platforms Multidisciplinary: chemistry + biology + physics + mathematics + informatics Untargeted: the metabolites are by definition not pre-defined Unfeasible validation: hundreds to thousands metabolites, many unknown Self-awareness: minimum reporting standard / levels of annotation Mass Spectrometry (MS) Direct infusion/Imaging Gas Chromatography (GC) Liquid Chromatography (LC) Capillary Electrophoresis (EC) Nuclear Magnetic Resonance (NMR) NMR: up to 100 metabolites few hundreds metabolites ESI- ESI+ Reverse Phase (RP) Normal Phase (NP) few hundreds metabolites few thousands metabolites GCxGC few hundreds metabolites Derivatisation No method can cover all metabolites Each method has advantages and disadvantages Certain overlap between the different methods exists
- 20. Metabolomics: no separation Mass Spectrometry (MS)Nuclear Magnetic Resonance (NMR) 1,5-dicaffeoylquinic_acid_pos m/z 50 100 150 200 250 300 350 400 450 500 550 600 % 0 100 1_5_dicaffeoylquinic_acid_pos 1144 (27.590) 2: TOF MS ES+ 1.16e3499.1252 163.0385 145.0289 89.0393 164.0447 337.0925 !;214.0880 319.0829 355.1047 ! 498.5406 539.1185 ! 540.1191
- 21. Metabolomics: no separation (extraction-solubility) Cyclohexane Petroleum ether Hexane Diethyl ether Chloroform Ethanol Acetone Methanol Water least polar most polar
- 23. Metabolomics: NMR Quan-Jun et al. RSC Advance 2015
- 24. Metabolomics: NMR Akhter et al. Anal Bional Chem 2016
- 25. Metabolomics: NMR Akhter et al. Anal Bional Chem 2016
- 26. Metabolomics: NMR Wei et al. Scientific Reports 2018
- 27. Metabolomics: NMR Wei et al. Scientific Reports 2018
- 29. Metabolomics: Mass Spectrometry Akhter et al. Anal Bional Chem 2016
- 30. Gaugg et al. J Breath Res 2016 Metabolomics: Mass Spectrometry
- 32. Metabolomics: Separation before detector
- 33. Metabolomics: Separation before detector
- 34. Metabolomics: Separation before detector
- 35. Metabolomics: Separation before detector Michael Witting, Metabolomics2018
- 37. Metabolomics: facts Holistic approach: complementary platforms Multidisciplinary: chemistry + biology + physics + mathematics + informatics Untargeted: the metabolites are by definition not pre-defined Unfeasible validation: hundreds to thousands metabolites, many unknown Self-awareness: minimum reporting standard / levels of annotation Mass Spectrometry (MS) Direct infusion/Imaging Gas Chromatography (GC) Liquid Chromatography (LC) Capillary Electrophoresis (EC) Nuclear Magnetic Resonance (NMR) NMR: up to 100 metabolites few hundreds metabolites ESI- ESI+ Reverse Phase (RP) Normal Phase (NP) few hundreds metabolites few thousands metabolites GCxGC few hundreds metabolites Derivatisation No method can cover all metabolites Each method has advantages and disadvantages Certain overlap between the different methods exists
- 38. Metabolomics: Techniques NMR MS (infusion) LC-MS GC-MS CE-MS Sensitivity * *** **** ***** ***** Theoretical plate - - *** **** ***** Dynamic range ~2 ~5 ~6 ~6 ~6 Robust ***** ** * ** * Reproducibility ***** * * * * Coverage * ** **** *** *** Velocity * ***** ** * *** User friendly * **** *** **** *** Data processing **** ***** **** * ** Identification ***** * ** **** * File dimension ***** ** * * ** Cost * *** **** ** *** Results * * ***** *** *
- 39. Metabolomics: MS analyzers Quadrupole (Linear) Ion Trap TOF Orbitarp ICT Resolution power (x103) 3-5 4-20 10-60 100-240 750-2500 Mass accuracy (ppm) low low 1-5 1-3 0.3-1 m/z range (upper limit, x103) 2-3 4-6 10-20 4 4-10 Acquisition speed (Hz) 2-10 2-10 10-50 1-5 0.5-2 Linear dynamic range 105-106 104-105 104-105 5x103 104 Price lower moderate moderate higher high Holcapek et al. JCA 2012
- 40. Metabolomics: workflow Experimental design Experiment Sampling LC-MS analysis Markers detection Markers validation Markers identification Hypothesis generation Method adaptation Metadata Randomi- zation Clear Question few variables Simple sample preparation Biological variability 60-150 samples Data processing XCMS QC OPLS-DA Visual control TargetLynx Statistics Targeted analysis Internal database External database MS/MS Arapitsas et al. JCA 2016
- 41. Metabolomics: experimental design - sampling Experimental design Experiment Sampling LC-MS analysis Markers detection Markers validation Markers identification Hypothesis generation Metadata Clear Question few variables Biological variability 60-150 samples Data processing untargeted vs. targeted approach two different tools development and validation analysis data analysis untargeted metabolomics targeted analysis
- 42. Experimental design Experiment Sampling LC-MS analysis Data processing Sumner et al. Metabolomics 2007 Metabolomics: experimental design - sampling Markers detection Markers validation Markers identification Hypothesis generation Minimum reporting standards Sample Preparation o Sampling process and protocol • Replicates (min 3 biological, proposed 5) • Harvesting/collection method • Sample processing (lyophilization, homogenization) • Storage (temperature, duration, atmosphere) • Relocation/Shipping
- 43. Experimental design Experiment Sampling LC-MS analysis Data processing Sumner et al. Metabolomics 2007 Metabolomics: experimental design - sampling Markers detection Markers validation Markers identification Hypothesis generation Minimum reporting standards Sample Preparation o Sampling process and protocol o Extraction protocol • Solvents, pH, buffer, temperature, time, volumes and weights (degassing) • Concentration, Dilution, Resolubilization • Enrichment (SPE), Desalting, Molecular weight cut off, Ion exchange • Clean up • Storage and/or Relocation
- 44. Experimental design Experiment Sampling LC-MS analysis Data processing Sumner et al. Metabolomics 2007 Metabolomics: experimental design - sampling Markers detection Markers validation Markers identification Hypothesis generation Minimum reporting standards Sample Preparation o Sampling process and protocol o Extraction protocol • Injector and oven temperature(s) • Mobile phase(s) composition • Flow rate • Gradient • pressure Chromatography-Separation o Instrument description (manufacturer, model, software) o Injection (auto injection, volume, wash cycle) o Column and pre-column (manufacturer, model, parameters) o Derivatization, Spiking (IS) o Separation method
- 45. Experimental design Experiment Sampling LC-MS analysis Data processing Sumner et al. Metabolomics 2007 Metabolomics: experimental design - sampling Markers detection Markers validation Markers identification Hypothesis generation Minimum reporting standards Sample Preparation o Sampling process and protocol o Extraction protocol Chromatography-Separation o Instrument description (manufacturer, model, software) o Injection (auto injection, volume, wash cycle) o Column and pre-column (manufacturer, model, parameters) o Derivatization, Spiking (IS) o Separation method Mass spectrometry o Instrument description (manufacturer, model, software) o Sample introduction (GC, LC, direct injection) o Ionization source (ESI, EI), polarity, voltages, vacuum, gases) o Mass Analyzer (TOF, FT-ICR) and acquisition mode (full scan, MSn) o Data acquisition parameters (scan range, calibration, accuracy, resolution, lock mass)
- 47. Experimental design Experiment Sampling LC-MS analysis Method adaptation Randomi- zation Simple sample preparation Data processing www.random.org/sequences/ Metabolomics: wet lab Markers detection Markers validation Markers identification Hypothesis generation Tip #3: Randomize your samples* *the instrumental analysis or better before the sample preparation Tip #2: Keep your sample preparation simple
- 48. Experimental design Experiment Sampling LC-MS analysis Data processing XCMS QC Metabolomics: dry lab – data processing Markers detection Markers validation Markers identification Hypothesis generation Tip #4: QC samples are your best friend* Tip #5: Understand/Know your instrument/method** Tip #6: Blacks and Standard mixes are not good QC sample Tip #7: Think if you really need IS(s) *show my your friend and I will show you your future **velocity, coverage, resolution, accuracy, robustness during the time of analysis machines are limited
- 49. Experimental design Experiment Sampling LC-MS analysis Data processing https://metlin.scripps.edu/xcms/ XCMS on-line XCMS metaXCMS XCMS2 mzMine OpenMS MetaboAnalyst Metabox Progenesis (nonLinear) MarkerLynx (Waters) MassHunder Profiler (Agilent) Compound Discoverer (Thermo) ChromaTOF (Leco) http://proteowizard.sourceforge.net/ XCMS Metabolomics: dry lab – data processing Markers detection Markers validation Markers identification Hypothesis generation https://xcmsonline.scripps.edu
- 50. Experimental design Experiment Sampling LC-MS analysis Data processing XCMS QC Metabolomics: dry lab – data processing Markers detection Markers validation Markers identification Hypothesis generation Tip #9: don’t underestimate peak picking and peak alignment
- 51. Experimental design Experiment Sampling LC-MS analysis Data processing XCMS Metabolomics: dry lab – data processing Markers detection Markers validation Markers identification Hypothesis generation Filling in missing peaks
- 52. Experimental design Experiment Sampling LC-MS analysis Data processing XCMS Metabolomics: dry lab – data processing Markers detection Markers validation Markers identification Hypothesis generation peak table variables samples
- 53. Metabolomics: markers Experimental design Experiment Sampling LC-MS analysis Markers detection Markers validation Data processing OPLS-DA Visual control TargetLynx Statistics Targeted analysis supervised multivariate analysis unsupervised multivariate analysis Markers identification Hypothesis generation
- 54. Markers detection Markers validation OPLS-DA Visual control TargetLynx Statistics Targeted analysis Markers identification Hypothesis generation Metabolomics: markers Experimental design Experiment Sampling LC-MS analysis Data processing supervised multivariate analysis unsupervised multivariate analysis -600 -400 -200 0 200 400 600 -800 -700 -600 -500 -400 -300 -200 -100 0 100 200 300 400 500 600 700 800 900 t[2] t[1] Scores Comp[1] vs. Comp[2] colored by Thesis Cellar House _H _C _H _C 01 02 02 19 10 20 _H _C 06 10 05 18 17 14 _C _H 01 04 17 08 15 18 _H _C 12 03 09 13 06 08 _C _H 05 20 16 04 16 19 _H _C 03 11 13 12 09 07 _C _H15 14 11 07 _H _C Hotelling’s T2 Ellipse (95%) = (774,3; 558,9) R2X[1] = 0,23 R2X[2] = 0,1198 EZinf o 2 - all QCs (M5: PCA-X) - 2013-04-22 11:48:48 (UTC+1) QC Cellar QC House Arapitsas et al. Metabolomics 2015; XCMS; Umetrics-Waters
- 55. Metabolomics: markers Experimental design Experiment Sampling LC-MS analysis Data processing supervised multivariate analysis unsupervised multivariate analysis Markers detection Markers validation OPLS-DA Visual control TargetLynx Statistics Targeted analysis Markers identification Hypothesis generation Arapitsas et al. Metabolomics 2015; XCMS; Umetrics-Waters
- 56. Metabolomics: markers Experimental design Experiment Sampling LC-MS analysis Data processing C T QC C T QC Markers detection Markers validation OPLS-DA Visual control TargetLynx Statistics Targeted analysis Markers identification Hypothesis generation
- 57. Experimental design Experiment Sampling LC-MS analysis Data processing Metabolomics: dry lab – data processing Markers detection Markers validation Markers identification Hypothesis generation Tip #10: don’t trust statistics – always turn to raw files
- 59. Metabolomics: identification/annotation Experimental design Experiment Sampling LC-MS analysis Data processing Markers detection Markers validation Markers identification Hypothesis generation Internal database External database MS/MS Try to identify everything the biomarkers (bio)markers 9737 Biochemical reactions Over 1000 types of reactions in Organic chemistry
- 62. Metabolomics: identification/annotation Experimental design Experiment Sampling LC-MS analysis Data processing Markers detection Markers validation Markers identification Hypothesis generation Tip #11: annotation needs time, be patient!
- 67. Progenesis QI