Quantification with Proteome Discoverer. Bernard Delanghe

Size: px

Start display at page:

Download "Quantification with Proteome Discoverer. Bernard Delanghe"

Lewis McKinney
6 years ago
Views:

1 Quantification with Proteome Discoverer Bernard Delanghe

2 Overview: Which approach to use? Proteome Discoverer Quantification Method What When to use Metabolic labeling SILAC Cell culture systems Small changes (10-50%) Peptide labeling li Dimethylation, ti TMT, Tissue proteins itraq Multiplexing (time courses) Moderate changes (20-200%) Label free using detector response Label free using spectral counting Single or Multiple Reaction Monitoring (SMR or MRM) 2 Extracted Ion chromatograms, area calculation Number of spectra per protein Absolute quantification, spiked with standards (AQUA) Many largely similar experiments Moderate changes (20-200%) Many highly similar experiments Large changes (>100%) Complex biological matrix (Serum) PinPoint

3 3 Proteomics and Quantification

4 New feature for Proteome Discoverer 1.2 Fast, Easy and automated Stable Isotope Precursor Ion Quantification Exploits the capabilities of high-resolution MS with precursor ion-based quantification 100k resolution 1. Classical SILAC with heavy Arginene and Lysine 2. SILAC with heavy Isoleucine 3. Peptides labeled with light, medium, and heavy dimethyl 4. Others like O16/O18, ICAT, ICPL 5. Doublets or Triplets 6. Using any enzyme 7. Any fragmentation technique (or a combinartion):cid, HCD, ETD, ECD 8. Any search engine (or a combinartion : Mascot, Sequest or Zcore 4

5 Preconfigured workflows Complete automation thru Xcalibur and Discoverer Daemon 5

6 Why are reliable and sensitive algorithms needed? Jesper_SILAC_HeLa #6382 RT: AV: 1 NL: 8.44E6 F: FTMS + p NSI Full ms [ ] Where are the SILAC pairs? Re elative Abundance Zoom -in m/z 6

7 Why are reliable and sensitive algorithms needed? Jesper_SILAC_HeLa #6382 RT: AV: 1 NL: 8.44E6 F: FTMS + p NSI Full ms [ ] Where are the SILAC pairs? ative Abundance Rel Zoom -in m/z

8 Why are reliable and sensitive algorithms needed? Jesper_SILAC_HeLa #6382 RT: AV: 1 NL: 2.76E6 F: FTMS + p NSI Full ms [ ] Where are the SILAC pairs? ative Abundance Rel Zoom -in m/z 8

9 Why are reliable and sensitive algorithms needed? Jesper_SILAC_HeLa #6382 RT: AV: 1 NL: 7.74E4 F: FTMS + p NSI Full ms [ ] Where are the SILAC pairs? Here ative Abundance Rel m/z 9

10 10 Result in Proteome Discoverer

11 Why are reliable and sensitive algorithms needed? Jesper_SILAC_HeLa #6382 RT: AV: 1 NL: 7.74E4 F: FTMS + p NSI Full ms [ ] ative Abundance Rel And here m/z 11

12 12 Result in Proteome Discoverer

13 Why are reliable and sensitive algorithms needed? Jesper_SILAC_HeLa #6382 RT: AV: 1 NL: 7.74E4 F: FTMS + p NSI Full ms [ ] ative Abundance Rel And here m/z 13

14 14 Result in Proteome Discoverer

15 Why are reliable and sensitive algorithms needed? Jesper_SILAC_HeLa #6382 RT: AV: 1 NL: 7.74E4 F: FTMS + p NSI Full ms [ ] Including the low abundant peptides and proteins Peptide intensity about 4000 counts ative Abundance Rel m/z 15

16 Precursor Quan: How Does It Work? Peptide Identification Event Detection Isotope Pattern Detection Protein Inference / Protein Grouping Isotope Pattern Multiplet Detection Peptide Quantification Peptide Classification Unique Redundant Not Unique etc. Protein Quantification 16

Event Detection Traditional TIC Chromatogram

Peak-detection on every good XIC Create an

17 Event Detection Traditional TIC Chromatogram Create XIC for every peak in every scan Keep XICs with shape of eluting component Event Chromatogram Event Chromatogram Peak-detection on every good XIC Create an event for each good good XIC Mass Intensity Area RT Etc. 17

18 18 Precursor Quan: Quan Result Display

19 Precursor Quan: Validation HeLa Digest (Arg10, Lys8) Mixture L : H = 1 : 3 Proteome Discoverer: Peptides at 1% FDR Log2 based Count % Median Mean * 68% Interval Unique Redundant Not Unique Peptides with Single Peak Channels Unique Redundant Not Unique Indistinguishable Channels Inconsistently Labeled Excluded by Method No Quan Values Total 2104 Quantified Not Quantifiable No Quan Values Processing Time [min] Proteome Discoverer Mascot Search 10 Event Detection 6 Quantification 2 Total 18 19

Precursor Quan: Validation HeLa Digest (Arg10, Lys6) Mixture L : H = 1 : 10 Proteome Discoverer: Peptides at 1% FDR Log2 based Count % Median Mean * 68% Interval Unique 422 31.14 8.51 8.49 1.58 5.

20 Precursor Quan: Validation HeLa Digest (Arg10, Lys6) Mixture L : H = 1 : 10 Proteome Discoverer: Peptides at 1% FDR Log2 based Count % Median Mean * 68% Interval Unique Redundant da Not Unique Peptides with Single Peak Channels Unique Redundant Not Unique Indistinguishable i Channels Inconsistently Labeled Excluded by Method No Quan Values Total 1355 Quantified Not Quantifiable No Quan Values >4 orders of magnitude Processing Time [min] Proteome Discoverer Mascot Search 8 Event Detection 1 Quantification 1 Total 10

21 Robustness and quality of the algorithms Number of raw files 72 Total file size (GB) 18.5 Total number of MS2 spectra Excellent variability! Summary of 3 replicates of 24 fractions of a Hela experiment (raw files provided with the Maxquant Software): > 190,000 peptides at 1% FDR 3560 protein groups; >12,000 proteins 96.5% of peptides quantified! 21

Stem Cell Kit #88200 Mouse Embryonic Stem

22 Complete SILAC Quantitation Workflow Pierce SILAC Quantitation Kits [500 ml media (2x), 50 ml dialyzed FBS (2x) and 50 mg of 13 C 6 L-lysine (1x), L-lysine (1x), L-arginine (2x)] RPMI 1640 kit, #89982 DMEM kit, #89983 Human Mesenchymal Stem Cell Kit #88200 Mouse Embryonic Stem Cell Kit #88206 LTQ Orbitrap Velos Proteome Discoverer 1.2 Comprehensive Quantitative Proteomics enabled by simultaneous MS and MS/MS 22

23 Precursor Quan: Workflow for Unlabeled Peptides Precursor Ions Quantifier node has area calculation included Can be used in conjunction with Reporter Ions Quantifier node 23

24 Spectral Counting Bernard Delanghe

25 Definition The total number of identified peptide sequences (peptide spectrum matches) for the protein, including those redundantly identified. Standard feature in Proteome Discoverer 25

26 Protein Area Calculation Alternative method in Proteome Discoverer: Protein area calculation Average of the 3 highest peptide areas Same algorithms as for Precursor Ion Quantification Standard workflow 26

27 Comparison of different quantification methods LTQ Orbitrap Velos raw file, SILAC labeled (Lys 8, Arg 10) HELA cells Fixed Heavy Light ratio 3/1 686 proteins identified (peptides at 1% FDR) Quantification methods SILAC: Heavy/Light ratio Area: Protein Area Heavy labeled peptide\ Protein Area Light labeled peptide (area calculated as average of the 3 highest peptide areas) Spectral Counting SILAC Area Spectral Counting Median Average SD % Quantified Proteins 82% 54% 51% 27

28 Distribution of Protein Heavy/Light ratios Top 100 Proteins Spectral counting as well as Peak area Is unreliable at lower protein abundance tio Heavy/Light Rat SILAC Heavy/Light Area Heavy/Light SC Heavy/Light Spectral counting is underestimating the ratios E E E E8 Area Light 28

29 Conclusion Proteome Discoverer has fast, robust, sensitive and reliable algorithms for Precursor based quantification. The workflow can be completely automated. Protein Area calculation method is clearly better than the spectral counting method. Thermo Fisher Scientific offers a complete workflow: from reagents to results. 29

Supporting Information. Lysine Propionylation to Boost Proteome Sequence. Coverage and Enable a Silent SILAC Strategy for

Supporting Information. Lysine Propionylation to Boost Proteome Sequence. Coverage and Enable a Silent SILAC Strategy for Supporting Information Lysine Propionylation to Boost Proteome Sequence Coverage and Enable a Silent SILAC Strategy for Relative Protein Quantification Christoph U. Schräder 1, Shaun Moore 1,2, Aaron A.