Data Integration, Mass Spectrometry Proteomics Software Development
Download as PPT, PDF3 likes1,216 views
This document discusses quantitative proteomics and integrating proteomics data into kinetic modeling in systems biology. It describes using isotopically labeled peptides to quantify proteins simultaneously via mass spectrometry. A method called QconCAT is outlined which uses an artificial protein containing multiple labeled peptides to reference. The document then describes an informatics pipeline to analyze such quantitative proteomics data, identifying peptides, determining concentrations, uploading data to repositories, and linking it with modeling databases to incorporate experimental data into systems biology simulations.
Data Integration, Mass Spectrometry Proteomics Software Development
- 2. Overview Quantitative proteomics Data integration in kinetic modelling in systems biology
- 3. A typical proteomics experiment Various routes through this map Separating by size or charge in most cases Identify peptides as a proxy for proteins, comparing theoretical to experimental spectra
- 4. Quantitative proteomics Approach described is qualitative Peptides / proteins identified but not quantified Mass spectrometry is not quantitative per se Different compounds have different physiochemical properties May ionise differently, more / less readily Therefore peak intensities cannot be compared between two different compounds Applies to peptides / proteins
- 5. Quantitative proteomics BUT peak intensities can be compared between compounds sharing the same physiochemical properties Isotopes Same physiochemical properties Different molecular masses (ΔM = 6Da)
- 7. Quantitative proteomics Can apply the same principle for peptides : IDVAVDSTGVFK IDVAVDSTGVF K* Lysine (K) residue is labelled with C 13 Same physiochemical properties Different molecular masses (ΔM = 6Da)
- 8. Quantitative proteomics Absolute quantitative proteomics requires isotopically-labelled peptide of known concentration spiked into sample Isotopically-identical peptides behave consistently Comparable peak intensity, comparable retention time Ratio of labelled over non-labelled peptide can be used to determined absolute concentration of sample peptide
- 9. Mixture 40:60 Data: Kathleen Carroll (Orbitrap MS) Quantitative proteomics: QconCAT
- 10. Quantitative proteomics: QconCAT Requirements: Determine absolute protein concentrations under a given cellular condition Quantify a number (~50) proteins simultaneously Apply QconCAT methodology Allows simultaneous introduction of many labelled peptides into sample Multiplexed absolute quantification for proteomics using concatenated signature peptides encoded by QconCAT genes. Pratt JM, et al. Nat Protoc . 2006, 1 :1029-43.
- 11. Quantitative proteomics: QconCAT Construct an artificial protein containing many peptides At least one from each protein of interest Ensure that the artificial protein is isotopically-labelled
- 12. Numerous absolute protein quantitations can be performed simultaneously Quantitative proteomics: QconCAT
- 13. … from instrument to browser From an QconCAT informatics perspective, there are three steps … Selection of QconCAT peptides Analysis and submission of data Browsing / querying
- 14. Selection of QconCAT peptides Q. Given a given protein, which peptides are suitable candidates for QconCAT peptides? Must… Be unique across organism Be detectable (digestible, flyable) Preferably… Be unmodified
- 15. QconCAT Selection Wizard Takes protein accession numbers as input (and other parameters) Provides list of potential QconCAT peptides Downloads sequence Performs BLAST against species-specific UniProt (tests uniqueness) Filters peptides “appropriately” Applies score to peptide, using PeptideSieve (predict flyability) Computational prediction of proteotypic peptides for quantitative proteomics. Mallick P, et al. Nat Biotechnol . 2007, 25 :125-31.
- 19. QconCAT… Multiplexed absolute quantification for proteomics using concatenated signature peptides encoded by QconCAT genes. Pratt JM, et al. Nature Protocols 1, 1029-1043 (2006)
- 20. QconCAT data analysis Identify and quantify peptides / proteins of interest Generate results in standard data format Facilitates data sharing Exploit existing software tools PRIDE XML PRoteomics IDEntifications Community developed standard http://www.ebi.ac.uk/pride/
- 21. QconCAT data analysis eXist database PRIDE XML Identify QconCAT Pride Wizard Quantify Format Upload Web / web service Browser Mascot PRIDE XML PRIDE Converter mzData
- 22. QconCAT data analysis eXist database PRIDE XML Identify QconCAT Pride Wizard Quantify Format Upload Web / web service Browser Mascot PRIDE XML PRIDE Converter mzData
- 23. Pride Converter Pride Converter (EBI) used to extract meta-data Who ran the sample, what was the sample, instrument used? etc. http://code.google.com/p/pride-converter/ PRIDE Converter: making proteomics data-sharing easy. Barsnes H, et al. Nat Biotechnol . 2009, 27 :598-9. Simple wizard allowing experimental data to be marked up with meta-data
- 24. Pride Converter
- 25. QconCAT data analysis eXist database PRIDE XML Identify QconCAT Pride Wizard Quantify Format Upload Web / web service Browser Mascot PRIDE XML PRIDE Converter mzData
- 26. QconCAT data analysis eXist database PRIDE XML Identify QconCAT Pride Wizard Quantify Format Upload Web / web service Browser Mascot PRIDE XML PRIDE Converter mzData
- 27. QconCAT data analysis eXist database PRIDE XML Identify QconCAT Pride Wizard Quantify Format Upload Web / web service Browser Mascot PRIDE XML PRIDE Converter mzData
- 28. QconCAT PrideWizard: Identify Goal: to identify heavily-labelled QconCAT peptides Uses Mascot http://www.matrixscience.com/search_form_select.html De facto standard database search engine for identifying peptides / proteins
- 30. QconCAT PrideWizard: Identify Mascot results are parsed to find labelled QconCAT peptides:
- 31. QconCAT data analysis eXist database PRIDE XML Identify QconCAT Pride Wizard Quantify Format Upload Web / web service Browser Mascot PRIDE XML PRIDE Converter mzData
- 32. QconCAT PrideWizard: Quantify Goal: to quantify heavily-labelled QconCAT peptides We now know m/z and retention time of peak identified as a QconCAT peptide First step: extract mass chromatogram for both heavy (labelled) and light (unlabelled) peptide
- 33. QconCAT PrideWizard: Quantify Extracted mass chromatograms Heavy and light peptide should overlay as they should have same retention time
- 34. QconCAT PrideWizard: Quantify Could use peak areas to quantify heavy versus light BUT hard (and inaccurate) to determine start and end
- 35. QconCAT PrideWizard: Quantify Alternative: extract individual scans showing isotopic clusters for both heavy and light
- 36. QconCAT PrideWizard: Quantify Apply sliding window and plot heavy versus light:
- 37. QconCAT PrideWizard: Quantify Final step: apply linear regression to determine heavy:light ratio (and an error ):
- 38. QconCAT data analysis eXist database PRIDE XML Identify QconCAT Pride Wizard Quantify Format Upload Web / web service Browser Mascot PRIDE XML PRIDE Converter mzData
- 39. QconCAT data analysis eXist database PRIDE XML Identify QconCAT Pride Wizard Quantify Format Upload Web / web service Browser Mascot PRIDE XML PRIDE Converter mzData
- 40. MCISB Proteome Database Searchable repository of quantitative proteomics data Geeky bit… eXist native XML database holding PRIDE XML JSP front end Querying extensible through XQuery Web and web-service interface Both human and computer-queryable
- 42. QconCAT informatics pipeline Reference: A QconCAT informatics pipeline for the analysis, visualization and sharing of absolute quantitative proteomics data. Swainston N, et al. Proteomics . 2011, 11 :329-33.
- 43. Data Integration
- 44. Systems biology modelling Enzyme kinetics Quantitative metabolomics Quantitative proteomics Systems Biology Model Parameters (K M , K cat ) Variables (metabolite, protein concentrations) PRIDE XML MeMo SABIO-RK Web service Web service Web service MeMo-RK Web service
- 45. Systems biology modelling Enzyme kinetics Quantitative metabolomics Quantitative proteomics Systems Biology Model Parameters (K M , K cat ) Variables (metabolite, protein concentrations) PRIDE XML MeMo SABIO-RK Web service Web service Web service MeMo-RK Web service
- 46. Systems biology modelling Enzyme kinetics Quantitative metabolomics Quantitative proteomics Systems Biology Model Parameters (K M , K cat ) Variables (metabolite, protein concentrations) PRIDE XML MeMo SABIO-RK Web service Web service Web service MeMo-RK Web service
- 47. Systems biology modelling Enzyme kinetics Quantitative metabolomics Quantitative proteomics Systems Biology Model Parameters (K M , K cat ) Variables (metabolite, protein concentrations) PRIDE XML MeMo SABIO-RK Web service Web service Web service MeMo-RK Web service
- 49. From experiment to simulation Kinetic models Experimental data Systematic integration of experimental data and models in systems biology. Li P, et al. BMC Bioinformatics . 2010, 11 :582.
- 50. Conclusion An informatics pipeline has been developed for analysis of quantitative proteomics data Data is associated with metadata , identified , quantified , and uploaded to database Community standards have been followed Experimental data can be incorporated in systems biology models Allows simulations of biological systems to be performed
- 51. Thanks…