Question: mzML parsing issue SpectraST 5.0 (stand alone, Linux, MacOS)

[Oliver Alka]

Hello everyone,

we are currently using a self-compiled version of SpectraST 5.0 (https://github.com/lukaszimmermann/SpectraST, with expat-2.0.1, gsl-1.14, zlib-1.2.5).

Creating a .sptxt from pepXML and mzML we get the following output:

Windows (correct): 

### SpectraST (version 5.0, TPP v5.0.0 Typhoon, Build 201610111222-7423 (Windows_NT-x86_64))
###
### IMPORT FROM PepXML "h:/HLA-Atlas/170407_AM_BD-ZH12_Lung_W_10%_DDA_#1new_400-650mz_msms50_idx_aligned.pepXML", (Comet against "sp_02_06_18_decoy.fasta" (AA); ) [P=0;q=9999;n=;g=FALSE;o=TRUE;I=CID-QTOF;_RNT=0;_RDR=100000;_DCN=0;_NAA=6;_NPK=10;_MDF=9999;_CEN=FALSE;_XAN=FALSE;_BRK=FALSE;_BRM=FALSE;_IRT=;_IRR=TRUE]
### ===
Name: IQPGAVKVY/2
LibID: 1
MW: 975.5742
PrecursorMZ: 487.7871
Status: Normal
FullName: L.IQPGAVKVY.A/2 (CID-QTOF)
Comment: AvePrecursorMz=488.0914 BinaryFileOffset=7042 CollisionEnergy=35.0 FracUnassigned=0.00,0/5;0.08,9/20;0.16,79/117 MassDiff=0.0000 Mods=0 NAA=9 NMC=1 NTT=0 Nreps=1/1 OrigMaxIntensity=3.8e+005 Parent=487.787 Pep=Non-tryp PrecursorIntensity=7.3e+005 Prob=0.0002 Protein=2/Q2UVX4/sp|P01024|CO3_HUMAN RawSpectrum=170407_AM_BD-ZH12_Lung_W_10%_DDA_#1new_400-650mz_msms50_aligned.10012.10012 RetentionTime=2759.2,2759.2,2759.2 Sample=1/h__HLA-Atlas_170407_AM_BD-ZH12_Lung_W_10__DDA__1new_400-650mz_msms50_idx_aligned_170407_AM_BD-ZH12_Lung_W_10%_DDA_#1new_400-650mz_msms50_aligned,1,1 Se=1^C1:pb=0.0002/0 Spec=Raw TotalIonCurrent=9.4e+005
NumPeaks: 117

RetentionTime=2759.2,2759.2,2759.2

Unix (not correct):

### SpectraST (version 5.0, STANDALONE)
###
### IMPORT FROM PepXML "current.pepXML", (Comet against "sp_02_06_18_decoy.fasta" (AA); ) [P=0;q=9999;n=;g=FALSE;o=TRUE;I=CID-QTOF;_RNT=0;_RDR=100000;_DCN=0;_NAA=6;_NPK=10;_MDF=9999;_CEN=FALSE;_XAN=FALSE;_BRK=FALSE;_BRM=FALSE;_IRT=;_IRR=FALSE]
### ===
Name: IQPGAVKVY/2
LibID: 0
MW: 975.5742
PrecursorMZ: 487.7871
Status: Normal
FullName: L.IQPGAVKVY.A/2 (CID-QTOF)
Comment: AvePrecursorMz=488.0914 BinaryFileOffset=378 CollisionEnergy=733.4 FracUnassigned=0.00,0/5;0.08,9/20;0.16,79/117 MassDiff=0.0000 Mods=0 NAA=9 NMC=1 NTT=0 Nreps=1/1 OrigMaxIntensity=3.8e+05 Parent=487.787 Pep=Non-tryp PrecursorIntensity=0 Prob=0.0002 Protein=2/Q2UVX4/sp|P01024|CO3_HUMAN RawSpectrum=170407_AM_BD-ZH12_Lung_W_10%_DDA_#1new_400-650mz_msms50_aligned.10012.10012 RetentionTime=487.8,487.8,487.8 Sample=1/_nfs_wsi_abi_scratch_leonb_HLA_Atlas_docker_leonb_test_170407_AM_BD-ZH12_Lung_W_10__DDA__1new_400-650mz_msms50_idx_aligned_170407_AM_BD-ZH12_Lung_W_10%_DDA_#1new_400-650mz_msms50_aligned,1,1 Se=1^C1:pb=0.0002/0 Spec=Raw TotalIonCurrent=4.9e+02
NumPeaks: 117

RetentionTime=487.8,487.8,487.8

I will use the retention time as an example, since some other entries such as BinaryFileOffset and CollisionEnergy seem to be not correct as well. 

The pepXML file is parsed correctly and reports the actual retention time (SpectraSTPepXMLLibImporter.cpp; line 360)
and writes the correct value in the SpectraSTPepXMLLibImporter::processSearchHit (line 412).

When checking the spectra via SpectraSTPepXMLLibImporter::readOneScan (line 1169), the rt gets overwritten,
with the one found in the Scan (line 1214), here the "scanInfo" is used to get the metadata from the mzML file. 

Unfortunatley it seems that in unix a wrong retention time was used (maybe due to a parsing error). 

I did not yet check further, but I wanted to ask if this came up before, or if anyone has an idea where to look (e.g. MzMLHandler)?
Could that be an issue of "our"/the standalone version or did anyone encounter something similar using TPP on a unix system? 

Thank you very much for your help!

Cheers
Oliver 
 

[oliveralka]

As an update: 

I was able to build TPP 5.0 and TPP 5.2 on Ubuntu 19.04. 

Using SpectraST with the above mentioned complied version was successful and everything parsed correctly. 

We are not yet sure what is going on with our version, but we are looking into it. 

[Eric Deutsch]

Thanks for reporting, Oliver. We have not seen this issue ourselves.

 

Regards,

Eric

--
You received this message because you are subscribed to the Google Groups "spctools-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to spctools-discu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/97ca3714-a74a-4048-894b-650e5bb439a0%40googlegroups.com.