Skip to content

Latest commit

 

History

History
74 lines (70 loc) · 7.18 KB

File metadata and controls

74 lines (70 loc) · 7.18 KB

Xisearch calculates a series of subscores for each CSM and combines (most of) these into the final score. The following table gives an overview of most of these

Explanations for most scores:

Score Description Source Search-Library Dependent Peptide1 Peptide2 Both Peptides   total Non-Lossy/primary Lossy Conservative Absolute Relative
Fragments
matched number of fragments matched Each fragment is only counted ones - no independent of how often it is seen FragmentCoverage   x x x   x x x x x x
unique matched same as "matched" but if a peak has more than one fragment matched to it, only one explanation is counted   x x x   x x x x x x
multi matched how many of the possible fragments where observed in more than one charge state   x x x         x   x
                           
fragment CCPepStubCount how many peptides are seen with crosslinker stub (0,1 or 2)         x              
fragment CCPepDoubletCount how many peptides are seen with crosslinker stub pairs (0,1 or 2)         x              
sequencetag coverage How much of the peptide is explained by consecutive fragments (minumum 3 consecutive fragments make a tag) SequenceTagCoverage   x x x             x
MaxCharge Maximum charge of any fragment matched FragmentChargeState       x           x x
MedianCharge Median Charge any fragment matched       x           x x
AverageCharge Average charge any fragment matched       x           x x
Spectrum
spectrum peaks explained how many of the peaks in the spectra are explained by the current match SpectraCoverage                     x
specrum intensity explained how much of the intensity is explained by the current match SpectraCoverage/SpectraCoverageConservative             x   x   x
spectrum Top10 matched % SpectraCoverage                     x
spectrum Top10 matched %                     x
spectra top20 matched%                     x
spectra top40 matched%                     x
spectra matched single% how much of the intensity all peaks without isotope cluster are explained                     x
spectra matched isotop% how much of the intensity all peaks with isotope cluster are explained                     x
spectra isotop% how much of the spectrum intensity is contained in isotop clusters                     x
Errors
Average MS2 error Error                   x x
Average1-RelativeMS2Error Avaerage of 1-(MS2Error/MaximalPermitedError)                     x
Precoursor Error                   x x
1-ErrorRelative 1-Precursor Error                     x
SearchLibraryScore
mgcAlpha Candidate Score for the alpha peptide -loge(Π(peptides matching peak/number off all fragments)) no group x x                  
mgcBeta Candidate Score for the alpha peptide (not used for candidate selecteion)     x x                  
mgxScore candidate score for the peptide pair     x   x                
mgxRank the ranking in the list of mgx-candidates     x     x              
mgcScore mgcAlpha+mgcBeta     x     x              
mgcDelta difference to the second best mgc score     x                    
FragmentLibraryScore similare to mgxscore but on whole spectrum instead of a denoised version FragmentLibraryScore x                    
FragmentLibraryScoreExponential scaleing of FragmentLibraryScore FragmentLibraryScore x                    
FragmentLibraryScoreLog scaleing of FragmentLibraryScore FragmentLibraryScore x                    
Other
SpectrumQualityScore combines several spectrum related scores to provide some inkling of how good the spectrum is matched and how good the spectrum itself is CombineScore                      
betaCount for the given alpha peptide how many beta peptides would match the gap-mass NoGroup                      
betaCountInverse 1/betaCount                      
Autovalidation Passes autovalidation                      
LinkSiteDelta how much better is the linksite then any other site (0 means its ambigious)                      
match score final score    x                    
delta diferrence to the secon best final score                      
                     
MatchScore to be ignored
J48ModeledManual001 to be ignored
AllScore to be ignored
MeanSquareRootError to be ignored
MeanSquareError to be ignored
Modified to be ignored
Containing to be ignored
Crosslinked to be ignored
BS3ReporterIonScore to be ignored

columns:

  • Source : where it is calculated (not really of interest here)
  • Search-Library Dependent : these scores depend on the fasta and modification settings. i.e. if you search the same data with a different e.g. a larger or smaller database even a match between the same peptides and the same spectrum will have somewhat different values.
  • Peptide1 : score exists as a version for peptide 1
  • Peptide2 : score exists as a version for peptide 2
  • Both Peptides: score exists as a version combining peptide 1 and 2
  • total: every ion is counted - independent whether it is a neutral loss or not. E.g. a peak matched with y5+loss of water would count as y5 seen.
  • Non-Lossy/primary: only basic ions (a,b,c,x,y,z) are counted - e.g. a peak matched to y5 counts as y5 seen, but y5+loss of water would not count
  • Lossy: coutns only how many fragments are seen as neutral loss. e.g. y5 does not count but y5+loss of water does
  • Conservative: every basic ion counts but also if a basic ion is not seen however three or more distinct versions with losses are seen count. E.g. if y5 is matched it counts but also y6 with 1 loss of water + y6 with 2 loss water and y6 with 3 loss of water seen would count as y6 seen.
  • Absolute: absolute values taken (think of errors +5 ppm error and +5 ppm error are considered the same.
  • relative: also taken as fraction of theoretical maximum. e.g. an error of 2ppm when searched with a maximum erro of 10 ppm would be 0.2 or seen 4 fragments matched for a 9 aminoacid long peptide would be 4/((9-1)*2)=0.25