novel method for estimating isotope incorporation using
play

Novel method for estimating isotope incorporation using the - PowerPoint PPT Presentation

Novel method for estimating isotope incorporation using the half-decimal place rule Ingo Fetzer Department of Environmental Microbiology userR Conference 2009, Rennes Problem 1.2e+7 100% 100% 0% 0% 50% 1.0e+7 Intensity 8.0e+6


  1. Novel method for estimating isotope incorporation using the ‘half-decimal place rule‘ Ingo Fetzer Department of Environmental Microbiology userR Conference 2009, Rennes

  2. Problem 1.2e+7 100% 100% 0% 0% 50% 1.0e+7 Intensity 8.0e+6 6.0e+6 4.0e+6 2.0e+6 0.0 835 840 845 850 855 860 865 870 875 880 mass � Substrate fluxes in Procaryotes � Function Activity Identitiy � Interactions: Competition Mutualism Page 2

  3. Goal Develop an algorithm for estimating 13 C incorporation by using ‘half decimal place rule’ Page 3

  4. ‘Half-decimal place rule’ (HDPR) Mann (1995) 100% Decimal places 0% m/z tryptic peptides Heliobacter pylori [Da] Schmidt et al 2003 Page 4

  5. Outline 1. Peptide mass calculation for 12 C and 13 C 2. Estimation of 12 C and 13 C slopes (HDPR) 3. Estimation of relative 13 C incorporation rates (of user data) implemented in ‘R‘ (R-project.org) Page 5

  6. Flowchart Script 1 Peptide database criterions Peptide sequences AA molecular formula Atomic weights Peptide sequence masses Page 6

  7. Peptide mass calculation for 12 C and 13 C Virtual digestion with MS-Digest dataset M. tuberculosis H37Rv amino acid sequences length 2 – 40 315,579 peptide fragments ChemScore ≥ 10 Missing cleavage = 0 Modifications = Null Mol. weight ≤ 5000 Da 90,637 peptide sequences Sanger Institute (ftp://ftp.sanger.ac.uk/pub/tb/sequences/TB.pep) Page 7

  8. Peptide mass calculation for 12 C and 13 C Page 8

  9. Peptide mass calculation for 12 C and 13 C Script 1: 1. Reduction of dataset (ChemScore, Modification etc.) 315,579 90,637 2 a. Peptide mass calculation Sequence + Molecular Sum of formula of in DB C, H, N, O for AA each sequence C 7 H 20 N 3 O 6 G=C 2 H 6 NO 2 A=C 3 H 8 NO 2 GAG Calculation of percentage 13 C Why? incorporation Page 9

  10. Peptide mass calculation for 12 C and 13 C 2b. Peptide mass calculation Sum of + Atomic weights Molecular weights of C, H, N, O of each sequences 12 C = 12.000000 Da sequence 13 C = 13.003355 Da (with decimal N = 14.003074 Da C 7 H 20 N 3 O 6 residuals) O = 15.994915 Da H = 1.007825 Da 12 C=242.135212 Da 13 C=249.158697 Da Page 10

  11. Peptide mass calculation for 12 C and 13 C 2a. Peptide mass calculation Atomic weights Count sum of Molecular + weights of C, H, N, O, S for 12 C = 12.000000 Da sequences each sequence 13 C = 13,003355 Da Molecular Masses of 12 C (with decimal N = 14.003074 Da O = 15.994915 Da residuals) and 13 C peptides H = 1.007825 Da S = 31.972071 Da H H O H H O O O C + 1 R C 2 R C C C C 1 R + C C H 3 O + OH OH + OH NH 3 N + NH 3 NH 3 + 2 R H M W (H 3 O + ) = 19.01839 Da Page 11

  12. Flowchart Script 2 Peptide sequence masses Transformation Transf. Peptide sequence masses k-means clustering Groupings Slope 0%/100% 13 C Linear fitting Transf. Decimal m/z – DR plot Residuals Page 12

  13. Estimation of 12 C and 13 C slopes Script 2: 1.0 0.8 Decimal residuals 0.6 m/z Temp = m/z - 1800 * DR 0.4 0.2 0.0 1000 2000 3000 4000 5000 6000 m/z Page 13

  14. Estimation of 12 C and 13 C slopes 1.0 Hartigan & Wong (1979) Decimal residuals 0.8 0.6 0.4 0.2 0.0 0 1000 2000 3000 4000 5000 6000 m/z Temp k-means clustering using kmeans() Page 14

  15. Estimation of 12 C and 13 C slopes 1.0 Hartigan & Add 0 1 2 3 Wong (1979) Decimal residuals 0.8 0.6 0.4 0.2 0.0 0 1000 2000 3000 4000 5000 6000 m/z Temp k-means clustering using kmeans() Page 15

  16. Estimation of 12 C and 13 C slopes 3.5 100% slope = 0.000633 0% slope = 0.000514 3.0 Decimal residuals 2.5 100% 2.0 1.5 0% 1.0 lm() function 0.5 Stats package 0.0 0 1000 2000 3000 4000 5000 m/z Page 16

  17. Flowchart Script 3 User data k-means clustering Robust Slope 0% + 100% 13 C linear fitting Slope user data reference calculation Estimation of relative 13 C incorporation rates Page 17

  18. Estimation of relative 13 C incorporation rates Script 3: 3.5 100% slope = 0.000633 0% slope = 0.000514 100% 3.0 Decimal residuals = ∗ DR b m/z (with a=0) 2.5 2.0 ‘outliers‘ 1.5 0% 1.0 Influence on slope estimation 0.5 0.0 0 1000 2000 3000 4000 5000 m/z Page 18

  19. Estimation of relative 13 C incorporation rates Linear fitting: 3.5 minimizing the error sum of the squares ∑ = − ∗ 2 min ( ) 3.0 SSE DR b m/z Decimal residuals 2.5 2.0 1.5 1.0 0.5 Breakdown point value = 0 0.0 0 1000 2000 3000 4000 5000 m/z Page 19

  20. Estimation of relative 13 C incorporation rates Robust linear model (Huber 1973) 3.5 M-estimator type Iterative re-weighted least squares (IWLS; 3.0 robust linear Decimal residuals Huber 1981, Hampel et al 1986) fitting 2.5 2.0 1.5 1.0 Breakdown point value set = 0.5 0.5 >50% of datapoints must change 0.0 0 1000 2000 3000 4000 5000 m/z rlm() function MASS package (Venable & Ripley 2002) Page 20

  21. Estimation of relative 13 C incorporation rates 3.5 100% slope = 0.000633 0% slope = 0.000514 100% 3.0 Decimal residuals = ∗ DR b m/z (with a=0) Your incorporation rate 2.5 is 98.3% 2.0 1.5 0% 1.0 − b b 13 = ∗ % 100 12 user C C 0.5 User − b b 13 12 C C 0.0 0 1000 2000 3000 4000 5000 m/z Page 21

  22. User data input R + Script 3 Page 22

  23. User data output R + Script 3 Page 23

  24. Sensitivity of Method Dataset Pseudomonas putida 1. Calculated 50% and 100% 13 C incorporation 2. Randomly sampled 100 times each 10-100 (steps by 10), 150, 200, 300, 500, 1000 sequences (0%,50% and 100%) 3. Statistics on estimated incorporation rate for 0%, 50% and 100% Page 24

  25. Sensitivity of Method 150 100% 140 ~100 samples 130 120 110 100 90 80 70 Incorporation [%] 60 50 100 50% 90 80 70 60 50 40 30 20 10 0 50 0% 40 30 20 10 0 -10 -20 -30 -40 -50 10 50 100 150 200 300 500 1000 Number of peptides Page 25

  26. Conclusion 1. ‚Half-decimal place rule‘ useful for the estimation of 13 C incorporation rates 2. Robust linear models better suited for fitting of highly variable user data than MinSSE fitting http://s3.amazonaws.com/readers/2008/08/14/draddalybig_1.jpg 3. >100 measurements needed for prescision <5% incorporation estimation Page 26

  27. Outlook 1. Application of HPDR on DNA 2. Backcalculation to 12 C-peaks function identification 3. Include N-isotope incoorporation http://www.sharp.co.jp/plasmacluster-tech/en/release/images/041117_3.gif Page 27

  28. Acknowledgement Nico Jemlich (UFZ) Carsten Vogt (UFZ) Martin von Bergen (UFZ) Hans-Hermann Richnow (UFZ) http://cache.gawker.com/assets/images/gizmodo/2009/01/bactsunsuet_01.jpg Hauke Harms (UFZ) Frank Schmidt (Uni Greifwald) Jens Mattow (MPI Berlin) Bernd Thiede (Uni Oslo) R development team Page 28

Recommend


More recommend