MSQuant is a tool for quantitative proteomics/mass spectrometry and processes spectra and LC runs to find quantitative information about proteins and peptides. Though automated it also allows manual inspection and change.
MSQuant’s main purpose is to make quantitation of proteins and peptides possible in the area of mass spectrometry/proteomics (science/analytical chemistry/molecular biology). E.g. an experiment where one heavy isotope-labelled sample from treated cells is compared to unlabelled, untreated cells to give quantitative answers about cellular processees. The input to MSQuant is a search result file (HTML) from the Mascot search engine (from Matrix Science) and one or more raw spectrum files.
MSQuant is written in Microsoft’s .NET and thus can only run on Windows computers that have the .NET runtime installed. The runtime is installed by default in later versions of Windows.
This program was directly referenced in the paper "A novel proteomic screen for peptide-protein interactions" (PDF) by Waltraud X. Schulze and Matthias Mann. 2003. Journal of Biological Chemistry (JBC). Abstract.
It was also the basis for the Nature paper "Proteomic characterization of the human centrosome by protein correlation profiling" (PDF) by Jens S. Andersen, Christopher J. Wilkinson, Thibault Mayor, Peter Mortensen, Erich A. Nigg, Matthias Mann. 2003. Abstract. Full text. On 2004-02-16 this paper was number 4 on Nature’s top 10 most downloaded papers.
SILAC is originally described in "Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics." (full text) by Ong SE, Blagoev B, Kratchmarova I, Kristensen DB, Steen H, Pandey A, Mann M. 2002. Abstract 1. Abstract 2.
SILAC is described in greater detail in "Properties of 13C-substituted arginine in stable isotope labeling by amino acids in cell culture (SILAC)." (full text NOT available online) by Ong SE, Kratchmarova I, Mann M. 2003. Abstract.
Triple encoding for SILAC is described in "Temporal analysis of phosphotyrosine-dependent signaling networks by quantitative proteomics." (full text as PDF file) by Blagoev B, Ong SE, Kratchmarova I, Mann M. 2004. Local copy. Abstract.
MS3 scoring is described in "Improved peptide identification in proteomics by two consecutive stages of mass spectrometric fragmentation" by Jesper V. Olsen and Matthias Mann. 2004. As PDF file. Abstract.
MSQuant is written in C# for .NET and in Visual Basic for .NET (VB.NET). It has an export facility to transfer information to databases that is supported by the .NET framework. For access to information in spectrum files it depends on vendor supplied COM components/libraries from (currently) three commercial packages, Sciex’s Analyst, Finnigan’s XCalibur and Micromass’ MassLynx.
MSQuant has now got a new website. Go to the new MSQuant home page.
2009-07-21. Note about the use of Mascot Result Launcher.
If you use one of the bundled applications in MSQuant, Mascot Result Launcher (MRL), to retrieve Mascot result files from the Mascot Server then you may get this message during parsing in MSQuant:
"The protein accession number could not be found. The Mascot result file (.htm/.html) does not have the correct format.
Possible reason: if you are using Mascot Result Launcher to get the result file be sure that the file "/usr/local/mascot/cgi/master_results.pl" on the Mascot server has been properly modified, in particular absolute (not relative) URLs in $pepViewLink, $protViewLink and $scoreGifLink. Contact your local Mascot server administrator with this information. Extra info: protein description: ..."
There are instructions on how to do this ("to use absolute URLs ") in a text file that is in the same folder as MRL. The name of the text file is "MRLdocs,for the Mascot UNIX adminstrator.txt" and this text file may be found by using menu "Utility/Open Helper Applications folder" in MSQuant, followed by opening of folder "MRL". In the text file look/search for the entry starting with "Edit file master_results.pl so $pepViewLink". This is not strictly necessary for parsing, but you may want to replace "mascot4.bmb.sdu.dk" with whatever is appropriate for your site.
2009-03-16. Patched version of MSQuant v1.5.
To overcome a conflict between the Finningan spectrum display component and DEP (Data Execution Protection) on Windows Vista both MSQuant and DTASuperCharge have had DEP disabled. The alternative to using these patched versions is to turn DEP off for the entire computer. In the future the Finnigan component may be fixed and/or an option added to MSQuant and DTASuperCharge to use another spectrum display component (e.g. the same display component as used for Micromass data).
Installation: copy each of the two files so they both overwrite the old version.
See also the notes below on how to run MSQuant and DTASuperCharge on Windows Vista.
Attention Windows Vista users. It is possible to run MSQuant and DTASuperCharge on Windows Vista - see the tips below. As of today there is also an update to the two programs that makes it easier.
Notes to Windows Vista users and others with restricted environments:
How to turn DEP off for the entire computer. Please only do this if you know exactly what you are doing - the computer may become inoperable otherwise. Turn off "Read-only" for file C:\boot.ini. Duplicate a boot line in file C:\boot.ini and change the text so you can distinguish two at boot time; e.g. add " DEP off.". Add this to the new boot line (or change of NoExecute if it already exists):
/NoExecute=AlwaysOff
Restart the computer and chose the DEP off line.
2009-02-19. DTASuperCharge, version 1.37 (1.3 MB) - an application for converting one or more Finnigan .RAW files to Mascot search input files in a format suitable for use with MSQuant.
Minor changes to tab order inside tab "DTA processing".
2009-01-05. For developers:
7-Zip package for MSQuant v. 1.5, 2008-12-19
(0.8 MB) with source code and other files.
Google indexable version.
2009-01-05. For developers:
7-Zip package for DTASuperCharge v. 1.31, 2008-12-15
(0.1 MB) with source code and other files.
2008-12-19.
Fifth official release
of MSQuant!!!. Version 1.5.
Added since b7: the target dialog is now opened on the first exports (e.g. exporting the parse report). All export menu commands can now use any target, e.g. to file or to OpenOffice. The dimension dialog bug was fixed (mentioned in b7).
Tips:
|
MSQuant helper applications
Supporting pre- and post-processing |
||||
|---|---|---|---|---|
| Name | Version | Date | Purpose | Notes |
| PILGrinder | 1.03 | 2009-02-12 | Simple mass calculations for peptides. | |
| DTASuperCharge | 1.33 | 2009-01-30 | Create Mascot input file from Finnigan LTQ-FT or LTQ-Orbitrap raw files. | |
| Mascot Result Launcher | 1.04 | 2008-11-11 | Efficient retrieval of the Mascot result file. Also contains a utility for converting non-Internet Exploiter 6 files to the format expected by MSQuant. | The on-demand formatting feature does not work if Mascot user access features are used on the Mascot server. |
| Mascot Result Viewer | N/A | 2007-12-17 | Extract a few specified proteins from the Mascot result file (to avoid waaaaaaiiiiiiting for the browser to show the whole file). | The tooltips do not work. Perl, e.g. ActivePerl, is required to be installed. |
| Finnigan Number Stripper | N/A | 2008-02-01 | Remove spectrum number fields from the Mascot result file. Allows use of MSQuant as a parse-only tool - the raw file is not accessed (and thus does not require any mass spectrometer software is be installed or can be used with mass spectrometers not supported by MSQuant). | Only works for the MSQuant formats - not for Mascot Distiller or Proteome Discoverer. Perl, e.g. ActivePerl, is required to be installed. |
| Mascot Result File Reductor | N/A | 2008-08-19 | Make the Mascot result file 5 times smaller... Can be crucial for parsing it at all. | Does not work for Mascot 2.1 result files, e.g. some of the sample files referenced in the Getting started section on this page. Perl, e.g. ActivePerl, is required to be installed. |
| MGFcombiner | 1.10 | 2008-09-26 | Merges MGF files and makes the MGF file suitable for MSQuant. | The underlying script is MultRawPrepare; referenced elsewhere on this page. Put MGF files in sub folder(s) under the specified input folder. This will name the output MGF file(s) after the sub folder name(s). The specified output file does not have any effect (an empty file is created...) and an extra MGF file, ".mgf" is generated. Perl, e.g. ActivePerl, is required to be installed. |
| MSRecal | 1.10 | 2008-07-31 | Recalibrates precursor masses, split into ID and non-ID peaklists. Greatly reduces input size for a second Mascot search - important for combined files. | Perl, e.g. ActivePerl, is required to be installed. |
2008-12-17.
Public
beta testing programme for MSQuant v. 1.5, beta 7:
1.5b7 (16.2 MB).
(Is in the 7-Zip format, not ZIP.
The 7-Zip software is
open source and an
installer is available).
48 hours remains of the beta testing programme.
Please report bugs as soon as possible.
Known bug: dimension dialog. If e.g. the size of a dimension is changed it is reset when button "More" or button "Fewer" is pressed.
Additions, changes and bug fixes since v. 1.5b3 (2008-12-10):Sequence coverage computation: various warnings are now displayed if there is no protein position information for peptides or if the protein sequence returned from the Mascot server is too short (due to outright incorrect sequence or to sequence database versioning issues).
After parse recalibration (re-recalibration?): was only working the first time... Not the second, third, etc.
Clicking on a point in the Recalibration Window (in MCR view) to open the Protein Validation window with the clicked peptide selected now works again. It was broken recently.
2008-12-10.
Public
beta testing programme for MSQuant v. 1.5, beta 3:
1.5b3 (16.2 MB).
(Is in the 7-Zip format, not ZIP.
The 7-Zip software is
open source and an
installer is available).
217 hours remains of the beta testing programme.
Please report bugs as soon as possible.
Additions, changes and bug fixes since v. 1.5b1 (2008-12-08):
Proteins can now be selected based on a list of accession numbers, from the clipboard. E.g. from ProteinCenter or from a spreadsheet (e.g. Open Office Calc or M$ Excel). If you have them in a file, first do this: open the file in a text editor (e.g. UltraEdit or NotePad), select all, copy.
2008-12-08.
Public
beta testing programme for MSQuant v. 1.5:
1.5b1 (16.1 MB).
(Is in the 7-Zip format, not ZIP.
The 7-Zip software is
open source and an
installer is available).
The beta testing programme will last 264
hours, starting from 2008-12-08T16:00:00Z.
Please report bugs as soon as possible.
Additions, changes and bug fixes since v. 1.5a61 (2008-10-22):
User defined protein tags. E.g. with: keratin, centrosomal, proteasomal, myosin, albumin and/or contaminant.
Columns in the Protein List window, Protein Validation window and Quantitation window can now be sorted without using the unhealthy rodent. Use menu View or corresponding keyboard shortcuts; Ctrl+1, Ctrl+2, etc.
It is now remembered which peptides have been used for recalibration. For now it appear in the export, but in the future versions it will be used in more useful places, e.g. in the recalibration vindow and in the peptide filter.
Added the reverse of Select All (proteins): Unselect All.
The blood pressure of Windows XP users can now be lowered even more: the recalibration visualisation dialog got parse/data set identification numbers in the window title: #1, #2, #3, etc.
Program default quant modes and modifications are now prefixed with "v2.1 ".
The almost trivial 1/1 columns are now narrow by default.
Bug fixed wrt. rodent-free computing: "Binned" and "ppb" in the recalibration visualisation dialog shared the same keyboard shortcut.
2008-11-11.
Extra MGF generator definitions
for Mascot Distiller, Proteome Discoverer, raw2msm and MaxQuant.
A new MaxQuant definition has been added.
Warning: the definitions still have not received a
lot of use/testing. Please report any
problems.
Tips:
2008-11-10.
CEBI's version of the MSQuant settings file "new_MSQ_quantitationModes.xml",
as of 2008-11-10:
Contains definitions for modifications and for
quantitation modes commonly used at CEBI.
In order to reduce confusion the name for most
Mascot 2.1 modifications is now prefixed with
"v2.1".
Tips:
2008-10-22. Note: there is a bug in Excel that affects MSQuant exports with PTM scoring (it may or may not have been fixed with Excel 2007). If you manually copy- paste between MSQuant exports opened directly by exporting from MSQuant, fields with more than 255 characters will truncated to 255 characters (e.g. PTM scoring information). There are several ways to work around this problem: 1. assure the export sheets are all opened in the same Excel "session" (process/ instance). E.g. close/save the directly exported sheets. Close down all Excel windows. Open Excel and open the saved documents from the File menu. 2. passing the information using the clipboard through a text editor as an intermediate (e.g. UltraEdit or NotePad). This will assure pure text and not the troublesome Excel clipboard format is used. 3. Use "Paste Special" when copy-pasting between Excel sheets. Chose "Text" and not default "Excel ..."
From http://ewbi.blogs.com/develops/2005/01/excel_s_text.html: "If you copy a cell (or more) with data exceeding 255 chr's from one work-book to another where the workbooks are in different Excel sessions [click on the excel shorcut instead of "New Document" or "Open..." within an existing session], the value are truncated to 255 characters."
2008-10-22. For developers:
7-Zip package for MSQuant v. 1.5a61, 2008-10-22
(0.8 MB) with source code and other files.
Google indexable version (no longer available).
2008-10-22.
Development (alfa) version of MSQuant:
1.5a61 (8.6 MB).
(Is in the 7-Zip format, not ZIP.
The 7-Zip software is
open source and an
installer is available).
The 3 scripts in the Utility menu are now integrated with MSQuant. They can be run on the selected Mascot result file in main window using the context menu (invoked by Shift+F10 - or the appropriate action using the unhealthy rodent). Thus you no longer need to edit BAT files in order to use these scripts - if the defaults parameters are acceptable. The scripts are: Mascot Result Viewer, Finnigan Number Stripper and Mascot Result File Reductor.
The "Highest scoring uniquely modified" option in the peptide filter dialog has been generalised. More options than uniquely modified are available.
The default values of parameters for spectrum reduction can now be set in settings file new_MiscOptions.xml (but they are still not saved when changing the values from the dialog in the Protein Validation window).
Improved parse report. a. Entries are now sorted by the key. It is no longer necessary to do this in the spreadsheet program (e.g. OpenOffice's Calc). b. The recognised modification section now lists both a line from the Mascot result file and the defined modification name in file "new_MSQ_quantitationModes.xml" (e.g. "Label:13C(6)15N(4) (R) " and "SILAC-13C615N4R ")
The warning message "No experiment point ..." when a raw file has not been mapped to an experiment point now explains what can be done about it. And there is only message box instead of one per missing raw file.
Empty fragment spectra are now handled more gracefully. Empty fragment spectra are caused by a bug in raw2msm (it repeats the previous fragment spectrum's peak list if the current fragment is empty and thus Mascot may identify a peptide for that empty fragment spectrum).
<none>
2008-10-07. Extra MGF generator definitions for Mascot Distiller, Proteome Discoverer and raw2msm. A new Mascot Distiller definition has been added; it extracts the retention time only and not the spectrum number and can thus be used in a parse-only scenario, e.g. MGF file generated from Agilent data. MSQuant 1.5a50 or later is required for this new definition to work.
2008-10-07.
Development (alfa) version of MSQuant:
1.5a50 (8.6 MB).
(Is in the 7-Zip format, not ZIP.
The 7-Zip software is
open source and an
installer is available).
A linear transformation can now be applied to the output value of all parse rules for the TITLE lines. This allows parsing of Mascot result files where the TITLE lines contain retention times in seconds (e.g. from Mascot Distiller).
Peptide filter dialog: SILAC modifications are now highlighted, can be checked by pressing a button and the checked state of all modifications can be toggled.
<none>
Raw file mapping dialog: the last entered raw file did not make it to the export (invoked by the button in the dialog).
2008-10-03.
Development (alfa) version of MSQuant:
1.5a49 (8.6 MB).
(Is in the 7-Zip format, not ZIP.
The 7-Zip software is
open source and an
installer is available).
consistency and other kinds of checks of the modifications and quantitation modes (defined in settings file "new_MSQ_quantitationModes.xml").
A warning message box will appear if only SIM spectra have been chosen for quantiation and the raw file has not been acquired with SIM spectra. The absence of this warning has probably confused a lot of users...
First version of an ability to describe the structure of the data/experiment, e.g. 6 sucrose fractions, each giving rise to 10 gel slices (and 60 acquired LC runs/raw files). Dialog "Experiment Dimensions" and "Experiment Raw File Mapping", menu Tools in the main window (blue).
New helper application added: Finnegan Number Stripper (script). Removes the spectrum number field and MSQuant can use the output file in a parse-only manner (no access to the raw file is performed and thus the vendor software need not be installed).
Export is now much faster. Sample: export of all 1267 proteins of a dataset took 17 seconds. Now it takes less than one second... In fact this improvement affects all functions that work on the protein selection. There is a fixed saving. E.g. 17 seconds for the sample proteins above. Post-parse recalibration visualisation and zapping intermediate quantitation information are now instant. The relative saving is less for the functions that take a long time: quantitation, MS3 scoring and PTM scoring.
Loading of a MB3 file is now much faster. It now takes 1-2 seconds. Previously it took 10-30 seconds.
The "Highest scoring uniquely modified" option in the Peptide filter dialog has been generalised. E.g. there can now be an identified peptide from ***each*** SILAC dish instead of only one (the highest scoring).
The detection of proper defintion/selection of SILAC modifications or quantitation mode can generate false alarms. The warning message has been changed accordingly.
Exports are now timestamped and also include the MSQuant version number and release date. This information is now the very first line in the export. Hopefully no scripts or other software will break as a result of this change...
The temporary files used by the export functions are now placed in the MSQuant installation folder instead of in the root of the C drive.
Helper application MRL: progress information is now displayed during conversion to MSQuant compatible file and it no longer appears frozen while it is retrieving the Mascot result file from the Mascot server.
Parsing the Mascot result file: the emPAI part is no longer included in the protein description.
Protein List window: other proteins than selected could be used. This bug affected protein/peptide export and post-parse recalibration/validation. Especially if the order of proteins was changed by e.g. sorting by protein mass. This bug was introduced on 2008-08-22.
Too many peptides made it through to the Quantitation window. All peptides with the non-SILAC modification set different from the non-SILAC modification set of the highest scoring peptide were added.
Zapped peptides were quantified on opening the Quantitation window. This could result in very long delays and even the belief that the application hang.
In helper application MRL: could run out of memory on large files, e.g. 300+ MB.
In the Quantitation window: the keyboard shortcut, F7, for Quantitation now works. There was another menu item using F7: menu View/Previous Protein.
2008-09-24. MSQuant and DTASuperCharge work on 64 bit Windows. Note: Contrary to what has been stated earlier it does work. There may be issues, but this is usually because the setup on the 64 bit Windows installation is different than on a 32 bit Windows installation. But you can just change the settings so it works...
2008-09-23. For developers:
7-Zip package for MSQuant v. 1.5a22, 2008-08-21
(0.7 MB) with source code and other files.
Google indexable version (no longer available).
2008-09-19. Extra MGF generator definitions for Mascot Distiller, Proteome Discoverer and raw2msm. Warning: they have not received a lot of use/testing. Please report any problems.
2008-08-28. CEBI's version of the MSQuant settings file "new_MSQ_quantitationModes.xml", as of 2008-06-26: Contains definitions for modifications and for quantitation modes commonly used at CEBI.
2008-08-21.
Development (alfa) version of MSQuant:
1.5a22 (8.7 MB).
(Is in the 7-Zip format, not ZIP.
The 7-Zip software is
open source and an
installer is available).
A number of helper applications, e.g.
DTASuperCharge, are now bundled with MSQuant.
Centroided MS spectra (often with only a single point
within the integration MCR interval) are now
handled appropriately during quantitation - but it
should be noted that centroiding of MS spectra
during acquisition is a bad idea...
Several helper applications are now bundled with MSQuant. They can be opened/launched from the "Utility" menu in the main (blue) window. See later entries on this page for details.
Added to the peptide filter dialog: a list of peptide modifications that must ***not*** be in a peptide. E.g. this allows peptides with the "Dioxidation (MW)" modification to be excluded from quantitation (by setting the peptide filter for validation appropriately either before or after parsing).
For all peptides in the Protein Validation (including unmodified peptides): the score (plus n and k) based on the displayed matched fragments is computed and displayed. It is only displayed, it is not remembered by the program; at least not in this version. It is intended for exploration and for manual verification of identification. For most higher scoring peptides the score is much higher than Mascot's score. Often more than 150 or 200.
More rules have been added to the new
generalised ion series. The spectrum type
(e.g. CID or ECD) can now also be matched. c
and z ions have also been added. Thus MSQuant
can now display and match the proper ions for
ETD spectra!
Sample screen-shot from MSQuant for BSA data
(0.002 MB).
For ETD: it works given a few gyrations;
Add a rule to spectrum classification (XML
settings file
"new_spectrumClassification.xml") to
properly classify ETD spectra (we are
reusing the ECD type here, but it is only
used internally):
<classificationItem>
<matchFirst>\@etd</matchFirst>
<matchSecond />
<mainClassification>enumIsFragmentSpectrum</mainClassification>
<MSsubType>enumIsFragmentECD</MSsubType>
<matchOrder>0</matchOrder>
<MSsubTypeAndFragmentationMechanismName>ETD MS2</MSsubTypeAndFragmentationMechanismName>
<dataSourceName>LTQ</dataSourceName>
</classificationItem>
It is now possible to specify only one modification allowed per residue in a given sequence. E.g. the same residue in a peptide cannot be both Phospho (ST) and IntactPhospho(STY). Sample: add this to modification "Phospho (ST)" to exclude 4 other phosphorylations:
<modificationsExcludedFromSamePosition>
<modID>5050</modID> <!-- v.2.1 IntactPhospho(STY) -->
<modID>5051</modID> <!-- v.2.1 phospho(STY) -->
<modID>5049</modID> <!-- v.2.1 -and 2.2?? Phospho (Y) -->
<modID>5547</modID> <!-- v.2.2 PhosphoIntact(STY) -->
</modificationsExcludedFromSamePosition>
Missed cleavages in the export.
MS3 spectra are now also matched to the phos-loss ion series (if the peptide has modification id 5048).
Empty regular expressions for the fields to parse from Mascot's title line (for each peptide) now means that some safe default value will be used. This allows e.g. Thermo/Finnigan Proteome Discoverer output to be used (that only have the spectrum number in it), given proper parsing rules are added to the XML settings file "new_MiscOptions.xml". Definitions for Proteome Discoverer and for raw2msm (0.002 MB). They must be inserted in the right place in settings file "new_MiscOptions.xml". E.g. right before the line with "</TITLElineParseDefinitions>"
The size of the export is now smaller as some low scoring PTM combinations are no longer output. This could result in too many combinations being output compared to the expected maximum of about 100. Sometimes this was too much for Excel to handle properly (32 K limit). In one example the number of combinations in the output was 334 and Excel would put them into several cells instead of just one; causing complete havoc. After the fix the number of combinations in the output was reduced to 151 and Excel was happy. In addition an absolute limit of 32000 bytes has been put in as the size could still exceed 32 KB for long peptides.
The windows titles have been changed so it is easier to navigate for Windows XP users... They now start with a parsing number and for some windows followed by the protein hit number. So all windows belonging to the same parse/data set now start with the same number.
Removed: the 'save' menu commands in Protein List window. Direct use of Excel has been eliminated in all export function in the Protein List windows' File menu. In this version it is somewhat confusing as the target (e.g. file or OpenOffice) is the last selected in the "Export Proteins and Peptides Information" menu command ...
Can now handle single point MS data in quantitation. E.g. if for some strange reason the MS scans for Finnigan data have been acquired as centroided. I have become aware of such data twice in the past few weeks. Screenshot of a quantitation window after this change.
"PIL ASSERT. aMaxY is undefined: 0". Only happend for unusual data, e.g. associating an unrelated raw file. Opening the Quantitation window. Also lots during quantitation.
For Micromass data: can now again open the Quantitation window. Symptom was: "Unhandled exception has occurred in your application. Values does not fall within the expected range"
2008-05-30.
Fourth official release
of MSQuant!!!. Version 1.4.3.
Added since b4: menu commands added to the Protein List window to open the containing folder for the Mascot Result file and the raw file.
2008-05-29.
Public
beta testing programme for MSQuant v. 1.4.3. Beta version 4:
1.4.3b4 ( 4.8 MB).
16 hours remain of the beta testing programme.
Please report bugs as soon as possible.
Additions, changes and bug fixes since v. 1.4.3b3 (2008-05-27):
Export functions work again and it can now be made to work on all systems - a very long standing bug that mysteryously made it not work on some systems is history... And it has been extended: 1. Direct export to Open Office's spreadsheet program, Calc. 2. Any application can be selected to open the output from the export functions. 3. If the export goes to a file then optionally it can be opened after being saved to disk using Windows's settings (Windows Explorer, menu Functions/Folder settings/ tab File Types/ Change), e.g. .TXT files opens in Open Office Calc or Excel (instead of Notepad or UltraEdit). MSQuant stores common install locations for Calc and Excel. If you have installed the applications in some other location then it can be made to work if you add the installation folder to the PATH (same way as for Analyst/QSTAR, described on this page).
2008-05-27. For developers:
7-Zip package for MSQuant v. 1.4.3b3
(0.7 MB) with source code and other files.
Google indexable version (no longer available).
Project MSQ1lib is now a C# project. Most C# source
files have been moved from the FastSerialisation
project to the MSQ1lib project.
2008-05-27.
Public
beta testing programme for MSQuant v. 1.4.3:
1.4.3b3 ( 5.0 MB).
The beta testing programme will last 72
hours, starting from 2008-05-27T13:00:00Z.
Please report bugs as soon as possible.
Known bug: the Export functions are broken, the corresponding Save functions must be used instead. An attempt will be made to fix this problem in the beta testing period.
Additions, changes and bug fixes since v. 1.4.3a74 (2008-03-07):Relocation of full file paths in MB3. MB3 files can now be moved to a different computer that has different drives and paths. This means you can also move Mascot result files and raw files to a different location without the fear that MB3 files will be useless.
Terminal modifications (e.g. O18 or Dimethyl N-term) are now supported in quantitation modes. Previously it only worked for computing correct fragment masses in the Protein Validation window. The correct masses are now computed and the majority of the peptides are no longer rejected during parsing.
Ion series has now been generalised instead of being fixed to b, y and y++. Any ion series can now easily be specified and the set to use can depend on some rules, e.g. on a modification, like phosphorylation. In this version things are hardcoded; if modification ID is 5048 ("Phospho (ST)" in standard CEBI version of settings file "new_MSQ_quantitationModes.xml") 8 extra ion series are used - y-98, y++-98, y- 196, y++-196, b-98, b++-98, b-196 and b++-196.
The dialog for save/export has been extended to include an option for spectrum line (LC point) information. This correspond to some old menu commands in the Quantitation windows, but you can now export this information for any protein subset from the Protein List window.
All OK buttons and Cancel buttons in all dialogs now have the standard keyboard shortcuts: Enter for OK and Esc for Cancel.
Now tolerant of old MB3 files with PTM scores of -Infinity. The Export/Save now succeeds. There are ASSERTs, but they have been greatly decreased in number.
All buttons in all dialogs now have an Alt keyboard shortcut. E.g. Alt+O for the OK button.
Protein Validation window: option to display unmatched ions or not. As a nice side-effect PTM (and MS3?) scoring is much faster when unmatched ions are not displayed.
Recalibration reality check. Dialog for entering e.g. minimum number of peptides for recalibration. This new feature also replaces the hardcoded limit of 4 for the minimum number of data points for recalibration. Started by a new button in the Options dialog: "Reality check..." Limitation: the settings are not (yet) remembered across program restarts.
Selection now works for the peptides in the Quantitation windows. E.g. you can now quantitate any subset of a protein's peptides. This replaces older menus command and buttons.
Matched ions in the Protein Validation window are now separated vertically from the unmatched ions.
All dialogs are now much closer to follow the standard: no resize handle, borders look like standard dialog borders, etc.
The Protein Validation window is taller now. This may or may not cause problems for some notebooks...
In Tools/Options dialog: "Preselected peptides" is now "PSPs".
the horisontal aligment for numbers in lists in the Protein List window, the Protein Validation window and the Quantitation window was changed from left alignment to right alignment.
In the Quantitation window: now always scrolls the spectrum list in the upper right so the spectrum line for the maximum LC point is visible (the spectrum data points are displayed in the lower left). This bug was probably very, very old.
All non-quantified peptides are no longer quantified on reopening the Quantitation window if one or more peptides have already been quantified. This resulted in unexpected long delay on opening the Quantitation window, perhaps even the belief that MSQuant had frozen.
Removed cause of this exception that could happen on exporting PTM scored peptides: "Could not export the selected XXX proteins". This problem exposed a problem deep down in the bowel of MSQuant. The implementation of "n over k" used in the computation of the PTM score (and MS3 score) could not handle n greater than about 170 and the result was a PTM score of Infinity or NaN ("not a number"); and later problems during export/save of PTM scored peptides.
ASSERT when displaying a peptide in Protein Validation window that pointed to the last fragment spectrum in a raw file. ASSERT message: "nextFragmentspectrum2() returned negative value - no fragment spectra past spectrum 10935".
the file that information is exported to is now closed by MSQuant when it is ready. Thus this file can now be deleted or renamed without having to resort to close MSQuant...
A problem with gi accession numbers and saving spectrum data points has been fixed. (The "|" from the gi accession numbers that was part of the proposed file name is not valid in a file name in M$ Windows)
Now gracefully handles if the number of (predicted) PTM combinations is greater than 2147483647.
Removed: harmless, but irritating and misleading ASSERT for some long peptides when displaying them in the Protein Validation window. (Limit is now 1200 instead of 500. For a sample peptide the count was about 800)
Is now compatible with raw2msm. If the start retention time output by raw2msm was 0.0 minutes then an ASSERT would occur.
2008-03-07.
Development (alfa) version of MSQuant:
1.4.3a74 (12.2 MB).
(Is in the 7-Zip format, not ZIP.
The 7-Zip software is
open source and an
installer is available).
This is primarily a bug-fix release.
LTQ-Orbitrap aware recalibration. Menu Tools/Options/ Recalibration of preselected peptides (precursor mass)/ Domain/Frequency, 1/f^0.5 (LTQ-Orbitrap).
New kind of plot added to recalibration Visualisation window: retention time vs. mass error. This makes it easier to spot if there was a mass drift during a run - set the peptide filter to narrow mass range to exclude any mass dependent mass error.
New unit in the Recalibration Visualisation window: ppb - part per billion. This makes it easier to read off values in the binning view as values are larger than one for most bins.
The Quantitation window finally got a distinct icon... And the Recalibration Visualisation window got one as well.
Proteins are now included in the protein list even if all peptides are low scoring (not in A,B,C). This corresponds to a different setting of a checkbox that was once in the Options dialog: "Do not show proteins without peptides in A, B or C". The protein score for those proteins is zero.
Now uses theoretical MCR instead of measured MCR in the Recalibration Visualisation window. This prevents unexpected sloped set of points for very high zoom states.
Now works again if there are no datapoints in the display mass range for the LC point to be quantified. The symptom was: "Arithmetic operation resulted in an overflow" This bug was introduced on 2008-01-31 with the theoretical isotopes markers. Workaround for older versions: exclude the peptide in question from the quantitation window somehow, e.g. clicking it off in the Protein Validation window (and saving to MB3 anew, etc if to be used from the BAT script).
Two problems related to peptide filter option "Highest scoring uniquely modidied" for the preselected peptides during parsing: a. The protein score is now correct. It would include peptide scores for the peptides excluded by the option. b. The peptide count in A,B,C (e.g. displayed in the Protein Validantion window and in the export) is now correct. It would include peptides that were excluded by the option.
Problem with the colours during zooming in the Recalibration Window. It was nearly impossible to see the zoom bounding box.
Combined files, menu Tools/Correlation Settings/Insertion of new peptides/Insert peptides only in range: the restriction on the retention time range for insertion of new peptides now works again. It was completely broken on 2007-01-08 (unless the range was specified in hours instead of minutes ...).
If the recalibration for a raw file can not be computed due to no peptides for recalibration (depends on peptide filter) then it is now effectively turned off; slope is set to 1.0 and offset to 0.0. The symptom was a very large number of message boxes with this text: "Calibration constants not found for key 811".
ASSERT during quantitation if a peptide was identified from a spectrum within 15 seconds of the end of the raw file (for the default -60s/+90s setting). This problem was thought to have been fixed on 2006-08-28. But it turned out only to be true if more than 15 seconds from the end of the file... Workaround for older versions: exclude the peptide in question from the quantitation window somehow, e.g. clicking it off in the Protein Validation window (and saving to MB3 anew, etc if to be used from the BAT script).
Combined files: now works if a predicted retention time happens to hit a region at the end of a spectrum file with only MS spectra... Symptom was: "PIL ASSERT. Spectrum number, 18981, is outside the range for the raw file, [1; 18980]". Note: these peptides are not inserted.
Quantitation window: menu File/"Export peptide with spectrum list" no longer causes an exception if the current peptide has not been quantified.
Recalibration Visualisation window. An exception is no longer thrown if the peptide filter is set such that all peptide MCRs are equal, e.g. only one peptide comes through the filter. The rather misleading error message was: "Could not read values dialog in dialog. Some numeric fields may hold letters, etc. Error: System.IndexOutOfRangeException: Index was outside the bounds of the array.".
2008-02-26. DTASuperCharge, version 1.19 (11.4 MB) - an application for converting one or more Finnigan .RAW files to Mascot search input files in a format suitable for use with MSQuant. Bug fixed: a trailing space is now output after the rawFile field (in the TITLE line). This bug made it more difficult to specify a robust parsing rule for MSQuant..
2008-02-11. For developers:
7-Zip package for MSQuant v. 1.4.3a59
(0.7 MB) with source code and other files.
Google indexable version (no longer available).
File PILinOut.vb was moved from
MSQuant/msquant/src/main/spcommon/ to
MSQall/MSQlib1/src/Utility/.
New C# source files have been put into the FastSerialisation
project/folder for the time being.
2008-02-11.
Development (alfa) version of MSQuant:
1.4.3a59 (4.3 MB).
(Is in the 7-Zip format, not ZIP.
The 7-Zip software is
open source and an
installer is available).
There is a lot of changes that will give different
results than previous versions, e.g. PTM scoring and
use of isotopes in quantitation.
The TITLElineParseDefinitions block in the settings
file new_MiscOptions.xml is now read by MSQuant and
the way information is extracted (from what is in
TITLE lines in the MGF file) is now thus
user-defined. Adding a new definition it should be possible to
directly parse files originating from other MGF file generators.
The main window (blue) can now be resized!!!
The Enter key can now be used to open spectra from quantitation spectrum list (in addition to double-clicking).
LTQ-FT aware recalibration (menu Options/Recalibration of preselected peptides/Domain/Frequency (LTQ-FT)). This will assign proper weights to the datapoints for recalibration; high MCR will not affect the recalibration result more than necessary.
Post-parse recalibration is now possible. Note: if you use a subset of the proteins to find the recalibration then the recalibration correction is still applied to ***all*** proteins. That is the reason for the exact same number of peptides in the Recalibration Visualisation window. If you want to review only the peptides that were used for recalibration you can use an existing function (menu Automation/Recalibration Visualisation. It works on the selection). Set the peptide filter in the Recalibration Visualisation window to the same as the filter in the options dialog. This is not an exact match as a peptide may be excluded if there are more than 300 available or if rejected in outlier elimination - but it is still useful.
New menu item for Protein List window: Report Raw Files Information. This makes it easier to evaluate the recalibration result for each raw file. E.g. the changes in ppm for low and high MCR are reported. Information about each file include: a. File without path. b. Internal code number: 1789, 809- c. Recalibration constants (for this raw file). d. After calibration value for the highest and lowest MCR. And corresponding absolute and relative differences to the two MCR values. e. Full file path.
The raw file ID (1789, 809-) has been added to the export.
Raw file filter to the peptide filter dialog. This is another tool for evaluation of the recalibration result for each raw file.
More default parsing rules to handle multiple raw files: 1. for QS 1.1 or later output 2. for newer versions of DTASuperCharge (assuming using a simple merge of MGF files - e.g. manual) that has a bug with missing trailing space. 3. for the newest versions of DTASuperCharge, 1.19, with the trailing space bug fixed. Thus for combined QSTAR files the output from Mascot.dll can be used directly! No scripts or other software is required. Warning: has only received minimal testing. To see the new rules you can rename settings file "new_MiscOptions.xml" and restart MSQuant. This will reset the very important quantitation window settings. Thus you should replace the blocks "quantitationSettings" and "PTMscoreSettings" in the new file with the content from the old file.
The recalibration trendline is now plotted (5 artificial points are added on this line to make it easier to spot non-linear lines). This line will only be linear if corresponding to the recalibration. E.g. it is not linear if unit is ppm and recalibration was done in the mass domain or if done in the frequency domain and plotted in Th.
Display of recalibrated peptides is now optional so that it does not have to be displayed every time. HOWTO: menu Tools/Options/button/"End of parse action..."
The theoretical isotope cluster heights are now indicated on the MS spectrum in the Quantitation window. It is the actual values for the current peptide, not just averagine. Current limitations: SILAC purity (e.g. 98.5%) is not taken into consideration and thus the values indicated for the non-wild isotope clusters are higher than they should be. The pre-peak height is not predicted either (but the program has been prepared for it). It does not work with zooming either.
Two new kinds of plots have been added to the lower right in the Quantitation window: a. LC profile visualisation using the same kind of plotting as for recalibration visualisation: the plot can be zoomed, panned, reset back to original scale, pop-up information for each data point, etc. It is read-only, but will hopefully in the future it will have the interactive features of the classic LC profile plot. b. Spectrum-to-spectrum SILAC ratios as a function of retention time. Makes it easy to spot trends caused by e.g. non-coeluting SILAC peptides.
File extension in opper-case in the Mascot result file is now accepted. E.g. a peptide with this information can now be succesfully parsed: rawFile: 20080117JEA_SILAC2_1_06.RAW
PTM-scoring is now different : n for PTM score was too high (and thus PTM score too low) due to a mass range set to the theoretical fragments mass range and not the mass range for the (observed) data. Example: For protein 1 in some data set: peptide GVMNAVNNVNNVIAAAFVK now get a PTM score of 80.1 instead of 72.8. The Mascot score is 88.0.
Now uses multiple isotopes for quantitation. In this version it includes the first isotope in the quantitation result. Future versions will be more sophisticated, including more isotopes depending on constraints like collision with the next SILAC isotope cluster group. The peak(s) used for quantitation are indicated in the MS spectrum in the Quantitation window. There will only be one if there is no signal for mono isotopic peak. This prevents spurious results if the isotope mass interval happens to pick up some (unrelated) data points.
Recalibration now has outlier-elimination. The algorithm used for this purpose is thought to be very robust and results are as expected sofar, but it remains to be critically tested (e.g. C13 masses with the previous implemented C13 correction turned off).
Recalibration is now done independently for each raw file (for merged files).
There is now a limit to the number of PTM combinations. This has been tested on an example that took 2 minutes to complete and on one that would have taken nearly 4 days (VSQTPIAAGTGPNFSLSDLESSSYYSMSPGAMRRSLPSTSSTS STK, 2oxM 3pST 1pY 6ipSTY 1hex). The limit for the number of combinations has a default value of 1000. It is specified through a field in settings file "new_MiscOptions.xml": maxCombinations in PTMscoreSettings. Sample:
<PTMscoreSettings>
<PTMminimumScore>14.000</PTMminimumScore>
<maxDisplayedPTMcombinations>25</maxDisplayedPTMcombinations>
<minimumPTMdeltaScore>0.1</minimumPTMdeltaScore>
<maxCombinations>18000</maxCombinations>
</PTMscoreSettings>
It is now much easier to zoom to the first and the last datapoint in a spectrum...
Recalibration itself is now somewhat more efficient with regard to memory and speed. It no longer go through all proteins and peptides. It now stops when the maximum of 300 peptides for recalibration has been reached.
Options dialog: a higher number of items for the quantitaiton mode can be seen at a time (20 instead of 8).
Minor corrections to the headers for the recalibration report (menu File/Export Recalibration Report).
More decimals were added for recalibration constants. Significant digits for FT data can be the 8 or more.
Minor changes in options dialog: size changes, item rearrangements, text changes.
Some menu items in the Protein List window got a new name.
When opening the containing folder for a Mascot result file or raw file (main window, context menu) the first existing containing folder is now opened if the file does not exist.
Added to default definitions for MGF file generator. A problem during parsing of the Mascot result with the output from DTASC and the option "Newer DTASuperCharge..." has been dealt with. The option was split into two and you can use the "... v.1.18 ..." for these kind of Mascot result files. The real problem is with DTASuperCharge, but MSQuant has now been adapted to work around the problem.
The progress display during parsing is back in business. It was broken on 2007-10-10. As a result the application is no longer frozen during parsing - you can select menu items or open another parse.
The MS3 spectra are again displayed as expected, not staying with the zoom for the MS2 spectrum.
In the export of quantified peptides from the Quantitation window the slope and offset recalibration constants were swopped.
Some fragment masses were computed incorrectly for peptides where the modification does not affect the entire peptide, e.g. terminal modification. For terminal modifications the masses of yN and bN were incorrect. As those masses were not used this bug was harmless. Symptom was an ASSERT during PTM scoring for peptides with terminal modifications.
Can now handle more than 8 phosphorylations in total... (e.g. 3 pST and 5 pSTY - or 11 pST). Such a peptide may take longer to PTM score than the remaining lifetime of our Universe, but that is a separate problem. Sample peptide: K.ALTSATIEDSMTQVMSSSRGPSPDQSTMSQDISTEVITR.L,9026-9064 + Deamidated (NQ); 11 Phospho (ST); 3 PhosphoIntact (STY); 2 Label:13C(6) (R) Symptom: attempting to open a protein where such a peptide would be displayed resulted in the infamous generic failure dialog: "... Problem opening the raw data file (wiff, raw or idx). Likely reasons: 1. raw file mode (Tools/Options) is Analyst or Micromass for a .raw file", with this specific message: "System.IndexOutOfRangeException: Index was outside the bounds of the array."
For QSTAR combined data sets the spectra for peptides can now again be displayed without this interruption: "PIL ASSERT. data_setWiffFileName() called more than once." This problem was introduced on 2007-08-08, v. 1.4.2b6.
2008-02-04. Updated:
N15 section.
Separate packages for Mascot 2.1 and for Mascot
2.2. With Mascot 2.2 it is much simpler to use N15
labeling with MSQuant.
2008-01-16. For developers:
Google indexable MSQuant source code (no longer available).
2007-10-11.
Development (alfa) version of MSQuant:
1.4.3a29 (11.2 MB).
(Is in the 7-Zip format, not ZIP.
The 7-Zip software is
open source and an
installer is available).
Version 1.4.3a24 was broken for QSTAR quantitation. This has been fixed.
In the recalibration visualisation window: median error in 100 Th mass windows. This makes it easy to spot any residual systematic error after recalibration or any systematic error before recalibration. Sample screen-shot (10 KB). (was reduced to 16 colours so the colour gradient is not exactly the same)
Quantitation is somewhat faster and somewhat more memory efficient (display update rate was decreased).
Quantitation now again works for QSTAR data.
The option "Only highest scoring uniquely modified" in the peptide filters now works again; peptides that should have been excluded could turn up in e.g. the export.
2007-10-03.
Development (alfa) version of MSQuant:
1.4.3a24 (no longer available) (11.2 MB).
Version 1.4.3a21 was broken for QSTAR data. This has been fixed.
Binning of the mass errors in the recalibration visualisation window. It includes information about what percentage of the displayed peptides will be included with more narrow mass accuracy settings in the Mascot search. If the peptide filter is set to display all peptides then it is also easy to see that the false positives are evenly distributed (in constrast to gaussian for the true positives). Sample screen-shot (13 KB). (was reduced to 16 colours so the colour gradient is not exactly the same)
none
Can now again open the Protein Validation window for QSTAR data.
2007-10-03. For developers:
7-Zip package for MSQuant v. 1.4.3a24
(0.6 MB) with source code and other files.
New files: SDUPkeyObserver.vb, AnalystSpectrumMarking.vb, overlaySpectrumMarking.vb, SDUPspectrumMarking.vb, ZedGraphSpectrumMarking.vb, frmRecalibrationVisualisation.Designer.vb, frmRecalibrationVisualisation.vb, frmZedGraphTryout.Designer.vb, frmZedGraphTryout.vb and frmZedSpectra.vb.
2007-09-27. CEBI's version of the MSQuant settings file "new_MSQ_quantitationModes.xml", updated for MSQuant 1.4.3a21 (terminal modifications and ICAT/HysTag mass differences): Contains definitions for modifications and for quantitation modes commonly used at CEBI.
2007-09-27.
Development (alfa) version of MSQuant:
1.4.3a21 (no longer available) (11.2 MB).
Good news for Micromass users: the dependence on the
Finnegan software has been eliminated. XCalibur is
no longer required to be installed in order to use
MSQuant with Qtof data!!
The problem with the C13 feature in Mascot has been
fixed.
Native support of terminal modifications. In constrast to
the work-around it no longer misses some
amino acids (e.g. MRKWP), the 30
modification site limit for PTM scoring is
much less of a problem and the number of
combinations for PTM scoring are much less
(higher speed and higher memory efficiency).
How to specify N-term modification for a peptide:
<startPosition>1</startPosition>
<endPosition>1</endPosition>
How to specify C-term modification for a peptide:
<startPosition>-1</startPosition>
<endPosition>-1</endPosition>
Automatic C13 correction: if the mass difference between the Mascot calculated mass (neutral) and the measured mass (+1 Da or +2 Da for C13 peptides) is greater than about +/- 0.5 Da then the measured mass (and corresponding measured MCR) is adjusted by the nominal mass difference times the C13- C12 negative mass difference, e.g. - 2.00670965 Da for a 2xC13 peptide. This may cause problems for recalibration if you have searched with a precursor mass accuracy of e.g. +/- 1.3 Da. In that case it is best to restrict peptides for recalibration to +/- 0.4 Da (or lower). (ppm value: compute for lowest precursor MCR.) Note that the measured mass and measured MCR for the C13 corrected peptides are ***permanently*** changed and no longer correspond to the values in the Mascot result file.
Y-values major gridlines for the recalibration visualisation window.
Now uses the open-source software Zedgraph to display Micromass spectra. There are a few cosmetic problems with the labels. This will be removed in a later version. Thus XCalibur is no longer required to be installed in order to use MSQuant with Qtof data.
Very important if you use HysTag or ICAT: the value of the field "massDiffFromBase" is now always that - a mass difference. Prior to this version MSQuant would use the value as an ***absolute*** residue mass if it was high enough (greater than 120 Da). The mass difference is relative to Cysteine native if it is a high value (HysTag and ICAT) and relative to Carbamido- Cysteine if it is a low value (e.g. N15). This distinction will eliminated when MSQuant is able to read the fixed modification from the Mascot result file and no longer needs to make assumptions about the modification for Cysteine.
If the raw file type is changed (e.g. from Qtof to LTQ-FT) then the quantitation window settings are now also changed. Previously the quantitation window settings would only be set at program startup.
For PTM scoring fragment spectrum display: the correct QSTAR spectrum is now displayed if it is not the first fragment spectrum, e.g. "experiment 3" or "experiment 4".
2007-09-08. For Qtof/Micromass/MassLynx and QSTAR/Sciex/Analyst QS 1.1 (or later) users: see the notes in the section "Combining several raw spectrum files" (close to the end of this page).
2007-09-07.
Development (alfa) version of MSQuant:
1.4.3a7 (11.2 MB).
The C13 problem can now be worked around, see below.
The built-in recalibration process is now visualised (sample). This makes e.g. the problem with the Mascot 2.2 C13 feature very apparent. It is displayed: a. At the end of the parse. For all peptides (in all proteins). b. From the Protein List window, menu Automation/Recalibration Visualisation. It works on the protein selection.
(Controls for recalibration visualisation: a. Resizing the window works! (when the window is not maximised). b. Rotate mouse wheel: zoom in / out. c. Drag out rectancle (with mouse): zoom to that area. d. Press the shift key. Cursor changes to a "hand" and panning is then possible. e. Hover the mouse over a point. Then the query number, peptide sequence, Mascot score, charge and protein number are displayed. f. Right click in spectrum/Undo All Zoom;Pan: returns to original plot. g. Right click in spectrum/Un-Zoom or Un-Pan: back to previous zoom/pan. h. The peptide filter can be used to restrict the plotted peptides. )
(Tips for recalibration visualisation: a. If you set the peptide filter to a minumum score of 30 then only the peptides used for recalibration will be displayed (if there are not too many). )
The set of peptides used for recalibration can now be set (through a peptide filter). An empty set will effectively turn off recalibration. The filter can also be used to prevent the C13 peptides from destroying recalibration (set a required mass accuracy that will exclude them, e.g. -/+ 500 ppm).
The standard shortcut key, Enter, corresponding to double-click now works in the peptide list in the Quantitation window.
A new field, "GUIdoShow", has been added to the quantitation mode definitions in settings file "new_MSQ_quantitationModes.xml" to enable/disable display of it in the Options dialog. Sample: <GUIdoShow>False</GUIdoShow> This makes for displaying only those that are actually used. This new field can added to an existing settings file. The same effect can achieved by simply deleting quant modes, but this way allows external tools to change it and in the future it may be possible to so also in MSQuant.
Keyboard shortcut for export...
Conditions for which peptides to use for
different things (e.g. validation during
parsing) have been generalised into a peptide
filter, taking items from 4 different
places (
peptide modification filter [menu Tool/Options],
peptide validation [menu Tool/Options],
peptide values filter [menu Tools/Peptide Filters] and
verified/quant filters
[Protein list window, menu File/Export Proteins;Peptides].
This much broader peptide filter is
available in the 4 different places as well
as in 2 new places: for recalibration and for
recalibration visualisation.
All peptide filters are remembered across program restarts.
Statistics are no longer available about how many peptides were not validated due to too high mass error.
A further condition for validating a protein during parsing is that at least one peptide must be validated.
In Tools/Options the label and possible values for what is different file formats (Analyst QS 1.1/2.0 vs. MSQuant classic) now instead refer to which software generates the MGF file. This is (hopefully) easier to understand for users.
The program default for the raw file type is now LTQ-FT (was QSTAR). This change is only important when MSQuant is initially installed.
If the file dialog for selecting a Mascot result or a raw file is dismissed with Enter then a parse (of some unrelated data set) is no longer initiated.
Saving (to a MB3 file) no longer fails if the "Remove Peptides for good" command has been used previously.
2007-08-20. Note: the new C13 feature in Mascot 2.2 ("# 13C") does not work at all with the current version of MSQuant (as the reported measured value is with C13 and the reported theoretical value is without). Recalibration will be completely wrong and thus the quantitation result will incorrect too. A new version of MSQuant that addresses this problem will be released soon (order of weeks).
2007-08-20. For developers:
7-Zip package for MSQuant v. 1.4.2 final
(1.8 MB) with source code and other files.
2007-08-10. CEBI's version of the MSQuant settings file "new_MSQ_quantitationModes.xml". Contains definitions for modifications and for quantitation modes commonly used at CEBI.
2007-08-10.
Third official release
of MSQuant!!!. Version 1.4.2.
Installation/usage notes:
2007-08-08. New beta version for MSQuant v. 1.4.2: 1.4.2b6 ( 0.3 MB). 53 hours remain of the beta testing period.
none
PTM scoring now work for QSTAR/Sciex/Analyst and Qtof/Micromass/MassLynx in all aspects (not only manually from the Protein Validation window). When started from the Protein List window and going to the next or previous PTM combination (from the Protein Validation window).
The Protein Validation window can now again be opened without error for Qtof/Micromass/MassLynx. This was broken with on 2007-07-11, corresponding to v. 1.4.2a23.
2007-08-08. For developers:
7-Zip package for MSQuant v. 1.4.2b5
(1.8 MB) with source code and other files.
New files: AnalystCOM.vb, QS11analystCOM.vb and QS20analystCOM.vb.
New approach for COM components (for Analyst only and for v.2.0 and v.1.1 in this version of MSQuant): now references several for each vendor, one for each significant version of the COM component(s). The corresponding interop DLLs are now created by a script ("Vendor COM versioning.bat") using Tlbimp and Aximp instead of using Visual Studio. Instead the generated Interop DLLs, with version numbers in the file name, are referenced directly in Visual Studio ( QS11.AxInterop.GraphControlLib.dll, QS11.Interop.ExploreDataObjects.dll, QS11.Interop.GraphControlLib.dll, QS20.AxInterop.GraphControlLib.dll, QS20.Interop.ExploreDataObjects.dll and QS20.Interop.GraphControlLib.dll. ). If you don't want or need a particular COM type or versions it is now relatively easy to leave them out as the references to them are much fewer than before (for now only true for QSTAR/Analyst). The dependencies to the Analyst COM components are now isolated to the derived classes of AnalystCOM.vb. This scheme will be extended to Analyst QS 1.0, Finnigan and Micromass, but not before 1.4.2 is released.
2007-08-03. New beta version for MSQuant v. 1.4.2: 1.4.2b5 ( 0.3 MB). 175 hours remain of the beta testing period.
To export: number of computed PTM combinations.
PTM scoring is now computed for modified peptides where all sites are modified (no doubt about the sites).
The parse rule for the elution time for QSTAR/QS 1.1 now matches any number of white spaces instead of a single white-space. Symptom was: "Problem extracting information from the tooltip section of the mascot result file"
During parsing it is now checked if a modified peptide contains one of the amino acids from the matched modification (this is sort of an incorrect definition of a modification). This, among other things, prevents problems later with PTM scoring.
During program startup it is now checked if the ID and if the match string of all defined modifications in settings file new_MSQ_quantitationModes.xml are unique. This (probably) prevents an "index out of range" exception during PTM scoring. (at this time it is not understood exactly why this is the case).
During program startup it is now checked if any of the prefix or postfix strings for the defined modifications in settings file new_MSQ_quantitationModes.xml contain uppercase letters. This prevents an ASSERT during PTM scoring.
2007-07-30. New beta version for MSQuant v. 1.4.2: 1.4.2b3 ( 0.3 MB). 260 hours remain of the beta testing period.
To export: wether a peptide is used in the protein quantitation result or not.
None.
For a reopened Quantitation Window: the protein quantitation values are now updated when a peptide is clicked on (or off). It used to work only in the initial Quantitation Window where the quantitation was done.
When a Quantitation Window is reopened: it is now indicated wether or not a peptide is used to compute protein quantitation values (before it was almost always on).
More of the scrollbar in the Protein Validation window is now visible (and therefore much easier to use...).
All of the third label in the Spectrum Reduction parameters dialog is now visible.
2007-07-26.
Public
beta testing programme for MSQuant v. 1.4.2:
1.4.2b1 ( 11.2 MB).
The beta testing programme will last 334
hours, starting from 2007-07-26T18:00:00Z.
Please report bugs as soon as possible.
Please note in particular the first mentioned bug
fix below...
Additions, changes and bug fixes since v. 1.4.1 (2007-03-06):
Support for Mascot 2.2 output. Please note: the default modifications (e.g. SILAC) in Mascot 2.2 are quite different from the ones previously used. Thus the file "new_MSQ_quantitationModes.xml" must be adjusted accordingly.
New feature to allow Analyst QS/QSTAR users to bypass running a script and use the output of Mascot.dll directly. Some parse rules are now defined outside the program, in the settings file "new_MiscOptions.xml" and a particular set (e.g. for later versions of Mascot.dll for Analyst QS) can be selected in the options dialog. It is the rules used to extract information from what was in the 'TITLE' lines in the input file for the Mascot, e.g. from "TITLE=Elution from: 31.16 to 31.16 period: 0 experiment: 4 cycles: 1". Note however that in this version definitions in "new_MiscOptions.xml" are for information only; at this point it is not possible to add new defintions and get it recognised by MSQuant. (May be implemented later: if you have software that creates TITLE lines with a different format than the MSQuant standard or later versions of Mascot.dll for Analyst QS then it may or may not be possible to add new defintions for it. The easiest way to get the definitions is to rename "new_MiscOptions.xml" and restart MSQuant. This creates a new "new_MiscOptions.xml" that you can then edit.)
Auto-doc (tm). Can now generate a report containing peptide information and its spectrum for all verified peptides of the selected proteins. It is intended for journal (e.g. MCP) requirements for documentation of raw spectra and matching ions. Output from MSQuant is in the DocBook format. This format can be converted into other formats, including HTML page(s) and PDF. Being a text format it also allows some postprocessing (e.g. reordering or leaving out information) before being converted to HTML or PDF. Preview of Auto-doc, HTML (for an early version of the Auto-doc feature). Preview of Auto-doc, PDF (for an early version of the Auto-doc feature). For your own data use the application "XMLmind FO Converter". Please refer to the "Tips" section above. To generate the DocBook files in MSQuant use the new menu command "Automation/Report..." from the Protein List Window.
In the Protein List window and the Quantitation window the columns with quantitation information can now be sorted...
For MS3 scoring: it is now possible to set the mass tolerance for the fragment ion matching (Protein Validation window, menu Protein/Peptide; Spectrum Reduction Parameters). This is also applied to PTM scoring and to the normal display of matching fragment ion peaks for a peptide. Note : the settings are not remembered by the program. They must also be set for each Protein Validation Window (they are not currently program wide settings).
Going one level down from the identification spectrum (to MS3) is now possible. (Protein Validation window, menu View/Down one spectrum level)
PTM scoring is now possible for non-Finnigan data.
PTM scoring can now work with amino acids in the peptide that are affected by more than one modification, e.g. pSTY and piSTY or terminal modifications work-around and any other modification. It has a new menu command, "PTM score - broader". The old PTM score is still available for reference.
The modification filter dialog can now handle a larger number of modifications.
Error reporting and error handling is now better if MSQuant runs out of memory while ***opening*** a Mascot result file for parsing (opening happens in the very beginning of the parsing). Example: after parsing "Secretome_d0_1_4.htm" (162 MB) for the first time MSQuant memory consumption is 482 MB. Parsing it for the second time may either result in an out-of- memory condition or in some cases it works (memory consumption 948 MB !!!). The out-of- memory condition resulted in this message: "PIL ASSERT. Select Case never fall-through.".
The "checked" state of a peptide in the Mascot Result file (.HTML) is now again parsed (from 2004-06-17/Mascot 2.0 until now it was considered always checked).
(Identified) SILAC pairs are no longer excluded from the Quantitation window. (in the same raw file, with the same charge and with the same modification state (mass)) Note: identified (by Mascot) SILAC pairs, not SILAC pairs in the MS data. (Fixed 2007-06-20.)
The options for including bold black peptides now work... They have never worked in the entire (or almost entire) life of MS Parser/MSQuant (!).
Now works for MS3 spectra and XCalibur 2 SR2 or later (XCalibur 2 SR1 worked OK; in SR2 two "cid"s are added to the spectrum headers - e.g. "ITMS + c NSI Full ms3 567.24@cid30.00 51808@cid30.00 [130-1050]"). The error message was misleading: "Unknown spectrum type, could not be derived from spectrum title (looking for ms3 or ms2) 'ITMS + p NSI d w Full ms3 948.07@cid30.00 1539.77@cid30.00 [410.00- 1550.00] '. The configuration file new_spectrumClassification.xml does not contain information to recognise the spectrum type, MS or MS/MS. Edit new_spectrumClassification.xml."
Recalibration now always represent the entire observed peptide mass range (in extreme cases, for very large number of peptides for the first protein, recalibration was broken). Example: the first 300 peptides have nearly the same mass, e.g. are only in the mass range [728.408539 - 733.419388 Th]. This can happen if the first protein has a very large number of peptides. MSQuant is now using (approximately) only the highest scoring peptide mass for each nominal mass; thus no particular mass or mass range has higher weight than others.
For a quantified peptide in the Quantitation Window: clicking off and clicking on did not have the expected behavior (e.g. protein ratios would stay at the ***clicked off*** values). This bug was probably introduced at 2006-10-10.
Some quantitation values in the export for a non-quantified peptide are no longer incorrectly using the value from some quantified peptide... Sample of this bug.
The menu command "Automation/Zap intermediate quantitation results" no longer deletes some ***peptide*** quantitation values (that it should not delete).
Fixed: display of the correct fragment spectrum for all versions of Analyst QS and for Mascot result files generated with the help of all versions of Mascot.dll.
Spectrum annotation (fragment ions) no longer disappears when going one level up and then one level down. (Protein Validation window, menu View/Down one spectrum level)
When there is no MS3 score the contribution is now correctly set back to zero. This can happen if MS3 scoring on different PTM combinations, e.g. first MS3 scoring on a combination where a fragment mass matches the MS3 precursor mass and then MS3 scoring for a combination where no fragment mass matches.
No longer crashes if regional settings is Norwegian (and possibly others)... Those where "Digit grouping symbol" is not "."
For MS3 scoring on PTM scored peptides: a bug was fixed that in rare cases made MSQuant crash (index out of range exception) or return incorrect results. It happend only for automation/MS3 and could not be reproduced by manual operations with MS3 score. It also depended on the order in which the MS3 scoring was done; even on which set of proteins was MS3 scored. This bug probably caused a mass difference of -42 Da to be used in some cases (depending on the defined modifications, etc.). This bug may have been introduced on 2006-08-01. Impact: low; the MS3 score could be incorrect in some cases, but only for a few peptides and only if PTM scoring was performed first and only if PTM scores were very low (below the display threshold).
If the name for the mb3 file to be saved contains a dot (".") then the resulting file on the disk will now always have the MB3 file extension. In the past it would not and therefore not show up in the dialog for opening a MB3 file.
2007-07-05.
Development (alfa) version of MSQuant:
1.4.2a23 (11.2 MB).
As from now on the software is placed in the 7-Zip format, not ZIP.
The 7-Zip software is
open source and an
installer is available.
Other archive software may be able to decompress the 7-Zip files.
I apologise for any inconvenience.
Installation: you must use the
MSQuant installer to install this version.
The unfortunate installer glitch of the previous
version has been fixed. The default tooltip format
after installation is now the most common, the
MSQuant standard format (not QS 1.1 or later). Plus
minor changes, some user options were removed.
2007-06-29.
Development (alfa) version of MSQuant:
1.4.2a22 (no longer available) (11.1 MB).
Installation (IMPORTANT !): You
***must*** use the installer to install this version
and due to an unfortunate installer glitch you must
also do this (even if you don't use QSTAR data):
make a copy of the file
"QS20.Interop.GraphControlLib.dll", rename the copy
to "GraphControlLib.dll" (check: the file size is 88
KB - it is easy confuse this file with the two other
files with similar names).
Should now truely really work (but remains to be
fully tested): display of the correct fragment spectrum
for all versions of Analyst QS and for Mascot result
files generated with the help of all versions of
Mascot.dll.
2007-06-21. Development (alfa) version of MSQuant: 1.4.2a19 (11.5 MB). Should now really work (but remains to be tested): can directly parse output Mascot result files that were generated with the help of the Mascot.dll for Analyst QS 2.0 (no longer needed to run a script). PTM scoring is now possible for non-Finnigan data (but remains to be tested). (Note: the "TITLElineParseDefinitions" section in settings file new_MiscOptions.xml is written, but in this version of MSQuant the content of that section is ignored when the program reads the file - instead the program defaults are used.)
2007-06-19. Development (alfa) version of MSQuant: 1.4.2a18 (11.5 MB) (no longer available). Now works for Analyst QS 2.0: can directly parse Mascot result files that were generated with the help of the Mascot.dll for 2.0 (no longer needed to run a script), the correct fragment spectra are always displayed and the installer installs the correct Interop files (Interop.ExploreDataObjects.dll). Note: the "TITLElineParseDefinitions" section in settings file new_MiscOptions.xml is written, but in this version of MSQuant the content of that section is ignored when the program reads the file - instead the program defaults are used.
2007-06-15. Updated: Qtof sample data. See section Getting started for details.
2007-06-06. Development (alfa) version of MSQuant: 1.4.2a14 (11.1 MB). Now again reads the content of settings file "new_MiscOptions.xml"... Particulary important for the quantitation window width for FT data !!! This was broken on 2007-05-02 - corresponding to 1.4.2a13.
2007-06-05. Note for Qtof: MSQuant does not (currently) work with MassLynx 4.1. For the time being you must use MassLynx 4.0 instead.
2007-05-18. Development (alfa) version of MSQuant: 1.4.2a13 (11.2 MB). Now works for MS3 spectra and XCalibur 2 SR2 or later (XCalibur 2 SR1 worked OK; in SR2 2 "cid"s are added to the spectrum headers). For MS3 scoring on PTM scored peptides: a bug was fixed that in rare cases made MSQuant crash (index out of range exception) or return incorrect results. Recalibration now always represent the entire observed peptide mass range (in extreme cases, for very large number of peptides for the first protein, recalibration was broken).
2007-05-18. For developers:
source code and other files for MSQuant v. 1.4.2a13 (1.0 MB).
For FastSerializer.cs please refer to the article
Optimizing Serialization in .NET - part 2.
The source code from that article can be used unchanged.
2007-05-16. DTASuperCharge, version 1.18 (10.6 MB) - an application for converting one or more Finnigan .RAW files to Mascot search input files in a format suitable for use with MSQuant. Now works for MS3 spectra and XCalibur 2 SR2 or later (XCalibur 2 SR1 worked OK; in SR2 2 "cid"s are added to the spectrum headers).
2007-05-16. For developers: Source code and other files for DTASuperCharge v. 1.18 (0.14 MB).
2007-05-11. Please note: MSQuant can only parse Mascot result files (.html) saved with Internet Exploiter 6 or earlier. It does not immediately work with these browsers: Internet Exploiter 7, Opera, FireFox, Netscape and others. There is however a work-around: a utility (.NET application) is available to convert a Mascot result file to the form of HTML that MSQuant expects. Use the button "Convert to MSQuant Readable File...". In the future MSQuant will be made browser agnostic.
2007-05-03. Note: if you use PTM scoring with MSQuant v.1.4.2a6 then prefix and postfix strings for modifications must not contain uppercase letters. This is a restriction, not a bug. This restriction may be lifted in a future version.
2007-04-24. Development (alfa) version of MSQuant: 1.4.2a6. Support for Mascot 2.2 output. A detailed description of other additions, changes and bug fixes in this version will be released with the first beta version of MSQuant 1.4.2.
2007-04-18. For developers: use of CVS has been discontinued. ZIP package for v. 1.4.1 final (0.9 MB): source code and other files.
2007-03-15. DTASuperCharge, version 1.17 (10.6 MB) - an application for converting one or more Finnigan .RAW files to Mascot search input files in a format suitable for use with MSQuant. Note 1: XCalibur must have been installed with the "XDK" option checked !!!!! Note 2: this application is dependent on some other application to generate input DTA files, e.g. extract_msn.exe (part of Bioworks). See below, section "Preparing input files for the Mascot search engine", subsection "LTQ-FT" for details.
2007-03-06. Second official release of MSQuant. Version 1.4.1.
2007-02-26. New beta version for MSQuant v. 1.4.1: 1.4.1b4.
Somewhat more memory efficient.
The protein score no longer includes peptides that are excluded due to the SILAC rule (fixed modification condition and from the same dish).
MSQuant can now really truely 100% handle more than 3 SILAC dishes. The headers for export of proteins lines are now correct; no longer fixed to triple encoding.
2007-02-26. The two major bugs have now been described, see the 2007-02-02 entry.
2007-02-15. MSQprojectFiles,2007-02-15b.zip. For MSQuant developers only. Updated sample project files for users of Visual Studio .NET, valid for version 1.4 final or later of MSQuant. Make sure to follow the instructions in MSQprjFIlesReadMe.txt (!!).
2007-02-07. There has been reports of trouble with upgrading to XCalibur 2.0 SR2, e.g. from XCalibur 2.0 SR1. Apparently the spectrum headers (for the same .RAW files) are now different. This has to be investigated. It may be necessary to change the content of settings file "new_spectrumClassification.xml".
2007-02-07. 121 hours remain of the beta testing period. No new bugs have been reported - so this version may be bug-free!
2007-02-02. Public beta testing programme for MSQuant v. 1.4.1: 1.4.1b2.
Compiled as 32 bit. MSQuant should now always run as a 32 bit application on a 64 bit operating system, e.g. "Windows XP Professional x64 Edition". This has not been tested, though.
Two (years old) major bugs in quantitation were fixed (they were somewhat dependent; the second masked the first):
1. Quantitation window: the threshold used
for the 20 percent rule is now local to
the (computer-) selected LC peak instead
of using the highest LC peak (global) in
the quantitation time window.
The 20 percent rule ensures that weak
spectra are left out of computations for
the peptide quantitation result.
Impact: no impact if the LC peak with the
MS- MS event was the highest peak in
the window. Otherwise the threshold
would be too high and some or all
spectra were ***not*** used for
quantitation; leading to no
quantitation result or to a result with
worse statistics than needed.
2. The checked state of spectra was not
consistent with what spectra were used for
computing the peptide quantitation result.
The state displayed was
before the 20 percent
rule on the LC peak was applied.
Impact: fewer spectra were used than
displayed. Very confusing for the user
and incorrect result compared to the
user selection.
No impact if the user
clicked or dragged the mouse in the LC
peak window in the lower right. In this
case the display and the internal values
became consistent.
For old versions you can verify this bug
by using menu File/"Export peptide with
spectrum list". With the new version you
can verify that it is now consistent...
Some minor bugs in quantitation were fixed.
For combined files, entry to the Quantitation window: some identified peptides are no longer excluded by inserted peptides.
2007-01-26. Version 1.4 of MSQuant has been withdrawn.
2007-01-11. First official release of MSQuant!! Version 1.4. (no longer available). This version includes the phosphorylation site scoring that was crucial to our Cell paper, "Global, In Vivo, and Site-Specific Phosphorylation Dynamics in Signaling Networks" (full text, PDF) by Jesper V. Olsen, Blagoy Blagoev, Florian Gnad, Boris Macek, Chanchal Kumar, Peter Mortensen, and Matthias Mann. 2006. Full text, HTML (may not be available). Abstract.
Any number of SILAC dishes for quantitation can now be handled; please note that column numbers in the export are now dynamic (are dependent on the number of SILAC dishes) and that some columns have been phased out; probably requiring changes to various scripts that process MSQuant export files.
PTM scoring - scoring of the modifications on different locations in the peptide. This works not only for phosphorylation (see above), but for any modification and can be used to get the correct fragment ion mass to be computed in the Protein Validation window. This is required if a peptide is only partially modified (e.g. one Methionine is oxidised according to Mascot, but there are several in the peptide sequence.)
The quantitation integration window (indicated by the blue lines in the lower left of the Quantitation Window) is now auto-centered on the MS-peak. This is useful when there is a significant space charge effect in full scan FT spectra (Finnigan data).
The quantitation signal thresholds are no longer hard-coded. They can now be defined by the user, currently by editing the file "new_MiscOptions.xml". The new fields are quantThreshold_QSTAR, quantThreshold_LTQFT and quantThreshold_Qtof but it is safest to let MSQuant generate a new version of this file: rename file "new_MiscOptions.xml", start MSQuant, exit MSQuant, edit the new file and transfer important settings/values from the old file.
Quantitation values for all isotopes are now displayed for each spectrum in the Quantitation window.
Retention time correlation can now be done at any time, optionally using the LC profile time centroid values instead of MS-MS event time, using any sub-set of the proteins. Use the new menu command Protein List Window/Automation/"Retention Time Correlation". You may want to first remove the existing inserted peptides using the new menu command Protein List Window/Automation/"Remove Peptides Inserted by Retention Time Correlation" (before or after quantitation).
Added option for only retaining the highest scoring unique modified peptides (main window, menu Tools/"Peptide filters").
Menu command to open the folder in which MSQuant is installed; main window, menu Utility/"Open MSQuant folder". For e.g. easy access to editing the setting files, "new_MSQ_quantitationModes.xml", "new_MSQ_correlationSettings.xml", "new_MiscOptions.xml" and "new_spectrumClassification.xml".
Find command, for the protein list. It can currently only search in the protein description. The found proteins are selected. To jump between the selected proteins use the new menu command: Edit/"Scroll to Next Selected".
More columns can now be sorted.
Prepared for ***specifying*** N for top N for export column PTM Info.
Saving and loading of the protein lists are now much faster, 10 times faster for saving and 20 times faster for loading. The loaded data in memory is also 4 times more compact than previously.
An average quantitation ratio value for a peptide is now actually the average of ratios (of spectra values) -not weighted by intensity. If you want to have it weighted by intensity then use the ratio of XICs (e.g. "3/1 (of XICs)" in the export.
The scaling of the quantitation result has changed. It is now the true area of the MS peak.
Most of the dialogs that were shown at the end of parsing are gone (e.g. "158 peptides were used for recalibration. Slope=0.9999817, offset=-0.0132602." and "Worst absolute error was 41.47 ppm. 74 peptides out of 200 (37.0 %) were un-validated because the error was greater than 10.00 ppm. ") - this information is now collected in a report: Protein List Window, menu File/Export Parse Report. In Open Office Calc or Excel: sort on column B. There is additional information; all the different modifications lines for peptides and how many modifications lines were recognised by MSQuant and how many were not recognised.
Definition of quantitation modes and modifications: the mass difference for amino acids in a modification need no longer be the same value. This allows N15 labeling to be used with MSQuant. See section "N15 labeling" at the very end of this document for further details.
Identical zero values are no longer checked when the mouse is dragged in the LC peak window in the lower right.
Some precursor ions can now also be precursors for MS3 scoring: precursors with progressive losses for the peptide modifications (currently only works for phosphorylation) - e.g. both 1 and 2 phos. (Or is it actually gains?? - check)
Retention time correlation: peptides are now considered unique when their modification set is different (before it was less precise, relying on a mass difference).
The value (single charged mass) for the lower limit for double charged y-ions was changed from 900.0 Da to 700.0 Da.
Binary save file extension changed to "mb3".
The assumed longest stretch of a spectrum file without fragment spectra was changed from 300 to 40000.
Value for detection of bad mass values changed from 100.0 Da to 120.0 Da in order to accomodate UBIK.
A number of bugs involving the LC profile for Finnigan data for some of the options "All MS spectra", "SIM spectra" or "Full scan spectra"; the symptom was an ASSERT, "... outside ... LC time range ... "
The full time range is now displayed in the LC profile at the lower right of the Quantitation, even if the LC intensities are identical zero at the extreme left or extreme right of the LC profile time windows.
The precursor mass (green) marked in the spectrum in the Protein Validation Window is now always computed correctly (!).
Some plot errors for the LC profile; points could be plotted outside the screen window and thus be invisible.
ASSERT when opening a protein whose peptides have been Zapped; was broken in 1.4.0a17, also another ASSERT for combined files was resolved.
Now again displays 0.0 for the protein ratio in the Protein List window when there are no checked spectra (no data for computing peptide/protein rations); it would keep displaying the last shown ratio for which there were spectra/data; probably broken.
External quantitation: the transformation from charged/observed mass to neutral is now correct.
Now also updates the protein ratio in the Quantitation window (in addition to the Protein List window) when the spectra are unclicked such that there are no clicked spectra to the upper right.
It is no longer possible to output comma as decimal separator to new_MiscOptions.xml - even if it is it is now detected.
2007-01-11. New section: Troubleshooting MSQuant installation.
2007-01-11.
Updated:
source code for MSQuant,
as of version 1.4.
For software developers only.
For FastSerializer.cs please refer to the article
Optimizing Serialization in .NET - part 2.
The source code from that article can be used unchanged.
2007-01-04. Documentation and scripts for using N15 labeling. with MSQuant has been released. For further details see section "N15 labeling" at the very end of this page.
2006-12-13.
Source code for DTASuperCharge
has been released.
For software developers only.
2006-05-18. First public release of DTASuperCharge - an application for converting one or more Finnigan .RAW files to Mascot search input files in a format suitable for use with MSQuant. XCalibur must have been installed with the "XDK" option checked!!!!! See below, section "Preparing input files for the Mascot search engine", subsection "LTQ-FT" for details.
2006-03-21. MSQuantInstaller,2006-03-21,1.4a16.zip Installer for version 1.4a16. (There is no official public release version yet).
Added: Filters for peptides, charge range and sequence length range for now. Sorting columns is now implemented for the most important columns in the three main types of windows: Protein List Window, Protein Validation Window and the Quantitation Window. For each protein: the computations for the protein ratio(s) are now also done for each raw file. New columns have been added to the export: "Raw file i/w", "Raw file i/w stdev", "Raw file h/w", "Raw file h/w stdev" and "Raw file, peptides". The raw-file file name is now always displayed in the Quantitation Window. It is now detected and reported if version 1.1 of Mascot.dll/IDA processor has been used without converting it to the 1.0 format that MSQuant understands (QSTAR/Sciex/Analyst only). All peptides for selected proteins can now be removed/deleted - this can be an advantage when there are memory problems. New columns with derived values have been added to the export: a. uncalibrated MCR error [Th], b. uncalibrated MCR relative error [ppm], c. absolute uncalibrated MCR relative error [ppm], d. calibrated MCR error [Th], e. calibrated MCR relative error [ppm], f. absolute calibrated MCR relative error [ppm], g. uncalibrated mass error [Da], h. uncalibrated mass relative error [ppm], i. absolute uncalibrated mass relative error [ppm], j. calibrated mass error [Da], k. calibrated mass relative error [ppm], l. absolute calibrated mass relative error [ppm].
Changes: The size of the Protein Validiation window is now more laptop friendly..... the buttons at the bottom of the window were sacrificed and corresponding commands with keyboard shortcuts can now be found in the menu Protein:Peptide; the buttons have instead been put into a separate toolbar (menu Protein:Peptide/Open Toolbar). Saved parses now have extension *.bs1. The XML format of "spectrumClassification.xml" has been changed and the new file name is "new_spectrumClassification.xml". Score for ZZZZ peptides is now left unchanged - is no longer set to 14.0
Bug fixes: Now again uses specified quantitation window (was incorrectly regarded as delta - bug introduced with 1.4.0a12). Now always displays the correct MS3 fragments when there is more than one possible MS2 precursor (e.g. "y++13" and "b7" for peptide "NQAALNPRNTVFDAK"). One or more fragments could be incorrectly highlighted - and the same number of fragments would be missing the highlight. Fix of a problem where the exported protein and peptide information (menu command "Export Proteins/ Peptides...") could be much too short; it would only export until the first inserted peptide (in the selected proteins); this bug was only present for: a. combined raw files, b. inserting new peptides is ON, c. using a version of MSQuant that supports flanking AAs (from Mascot 2.1)- that's all new MSQ versions after 2005-10-19, approx. final version of 1.4a14 and later. A bug in sorting the peptides was fixed -the impact of this bug is not clear. Now uses uncalibrated MCR values instead of calibrated ones when quantitating in the raw data. A bug was fixed for Micromass data, returning the mass range for the spectrum and the minimum and maximum signal in the spectrum.
2006-03-14. Added installation instructions for Analyst 1.0 QS Users, see section "Installation instructions".
2005-12-20. Notes to the QSbroken2QSclassic script.
2005-11-26.
QSbroken2QSclassic_script,2005-11-26.zip.
This script was itself broken... Now outputs "experiment number" as
expected by MSQuant -
this number is offset by one in Mascot.dll for QS 1.1 compared
to Mascot.dll for QS 1.0.
QSbroken2QSclassic is a Perl script to convert
output from the Mascot.dll in Analyst QS 1.1 to the
old format that MSQuant understands.
Perl must be installed on your
computer in order for this to work. For Windows
computers we can recommend ActivePerl,
http://www.activestate.com/activeperl/?psbx=1.
The exact version number does not matter.
2005-10-31.
MSQuantInstaller,2005-10-31,1.4a14.zip.
Installer for version 1.4a14.
(There is no official public release version yet).
Bug fixes:
Removed blocking ASSERT in quantitation.....
2005-10-28.
MSQuantInstaller,2005-10-28,1.4a13.zip (no longer available).
Installer for version 1.4a13.
(There is no official public release version yet).
Added:
In order for the program to parse the result
file a space is no longer required after the
content of the "Cycle:" field in the tooltip
line in the Mascot result file (this makes it
easier for script writers...). The MS3 score is
now identical zero and not a low value (e.g.
2.12745) if there is no evidence (matching
fragments).
Changes:
the two flanking amino acids are now scanned in
from the result file. For combined files: the
user is now warned if new peptides are inserted
and parenthesised is turned off. Is now robust
enough to behave well for empty MS3 spectra -
encountered when using older LTQ-FT data with
the very buggy XCalibur 2.0.
Bug fixes:
problem with beginning of LC window for
quantitation if it initially happened to hit a
MS3 spectrum for which the MS spectrum is a MS2
spectrum - now seeks the MS2’s MS spectrum in
that case; symptom was: "PIL ASSERT. Spectrum
number xyz is not the same as in the LC profile
(abc)...". Quantitation again works for "No
isotope mode".
2005-09-27.
QSbroken2QSclassic_script,2005-09-27.zip (no longer available).
Perl script to convert output from the Mascot.dll in Analyst QS 1.1 to
the old format that MSQuant understands.
Perl must be installed on your
computer in order for this to work. For Windows
computers we can recommend ActivePerl,
http://www.activestate.com/activeperl/?psbx=1.
The exact version number does not matter.
2005-09-27.
MSQuantInstaller,2005-09-27,develop.zip.
Development version of MSQuant for evaluation of the above script.
Note: should only be used for evaluation.
Fixes: the Protein Validation window again works with
QSTAR data...
2005-08-30.
MSQuantInstaller,2005-08-30,1.4a12.zip (no longer available).
Installer for version 1.4a12.
(There is no official public release version yet).
Added: now accepts the file name extension
.html (in addition to .htm). Shortcuts for
Accepting/Rejecting Peptides in the Protein
Validation window. Spectrum browsing feature in
the Protein Validation window: move up from the
spectrum that identified a peptide, e.g. up
from MS3 spectrum to MS2 containing neutral
loss, then to SIM spectrum and finally full
scan spectrum (currently only for Finnigan and
only for moving up); one or two
phosphorylations are marked in those MS
spectra; isotopes +1 and +2 are marked for all
monoisotopic peaks. Added report for the
recalibration: menu File/ Export/Save
Recalibration report in the Protein List
window. Old configuration files are no longer
tolerated in the folder from which the program
starts. Mass window for quantitation can now be
a function of the mass; the default in the
configuration file new_MiscOptions.xml is to be
proportional to the square of the mass for
Finnigan data. Keyboard shortcuts for Accept
Peptide and Reject Peptide in the Protein
Validation window. Multiple amino acids for a
modification is now supported - important for
correct computation of fragment masses, e.g.
phosphorylation, S and T. Fragment labels now
indicate the amino acid that is different
compared to lower mass fragment of the same
type. Spectra are now auto-matched, with fixed
parameters SmartPicking level 9 and Mann
reduction 100/4.
Changes: Proteins containing no peptides are
now excluded from the protein list (primarily
because of intenal errors). No correlation
between raw files takes place if insertion of
peptides is turned off by the user - speeds up
parsing and requires less memory. Some
theoretical masses now have a higher precision,
10 instead of 8 significant digits: Cysteine
and C-terminus (-OH).
Bug fixes: HysTag like modes are no longer broken.
Fragment masses for peptides with HysTag
modifications are now computed correctly (but
masses in the external XML file needs to
absolute; effective base mass must be 0.0 Da).
Fixed assert in handling of inserting peptides
- was broken bacause of changes related to
generalised quantitation modes. Computation of
fragmentation masses for phospho peptides is no
longer disturbed by fix for correct mass
computation for HysTag like modes (fix
introduced in this version, on 2005-07-07).
Zooming in spectrum in the Quantitation window
is no longer broken. HysTag masses for use in
the Protein Validation window fragment masses
are now computed correctly.
2005-08-30. MSQprojectFiles,2005-08-30.zip. For MSQuant developers only. Updated sample project files for users of Visual Studio .NET, valid for version 1.4a12 of MSQuant.
2005-06-13.
MSQuantInstaller,2005-06-13,1.4a11.zip.
Installer for version 1.4a11.
(There is no official public release version yet).
Bug fixes:
quantitation is no longer broken for non-triple encoding modes...
2005-06-12.
Version 1.4a10 - was never released!.
Added: Support of Mascot 2.1 result files.
Bug fixes:
an exception is no longer thrown during parsing
if the (new) quantitation modes are read from a file...
2005-06-03.
MSQuantInstaller,2005-06-03,1.4a9.zip (no longer available).
Installer for version 1.4a9.
(There is no official public release version yet).
Note: Please note that using current versions
of MSQuant for Micromass data it is
dependent on a component in the Finnigan XCalibur
software package for displaying spectra and thus only run
if XCalibur software package is installed.
Note: requires Microsoft .NET version 1.1.
Added: <to be added>
Changes: <to be added>
Bug fixes: <to be added>
2005-05-30. MSQprojectFiles,2005-05-30.zip. Note: the project files are ahead of time for the currently checked in source files in CVS. For MSQuant developers only. Updated sample project files for users of Visual Studio .NET.
2005-02-17.
MSQuantInstaller,2005-02-17,1.4a8.zip.
Installer for version 1.4a8.
(There is no official public release version yet).
Note: Please note that using current versions
of MSQuant for Micromass data it is
dependent on a component in the Finnigan XCalibur
software package for displaying spectra and thus only run
if XCalibur software package is installed.
Note: requires Microsoft .NET version 1.1.
Added: <to be added>
Changes: <to be added>
Bug fixes: <to be added>
2005-02-17. MSQprojectFiles,2005-02-17.zip. For MSQuant developers only. Updated sample project files for users of Visual Studio .NET.
2004-12-13.
Note: the QSTAR/Analyst software does not install or
work on Microsoft Windows XP, Home Edition.
Thus you can not use MSQuant to analyse QSTAR
data on that version of Microsoft’s operating
system. The Professional Edition of Windows XP
may or may not work. The safe choice is to use
Microsoft Windows 2000.
We have not experienced problems with
LTQ-FT or Qtof.
2004-11-12. Sample data set for Sciex QSTAR. See section Getting started for details.
2004-11-12.
MultRawPrepare,2004-11-12.zip.
Perl script for preparing Mascot input files such
that multiple raw spectrum files can be handled in MSQuant.
This script has successfully been in use by our
group for more than one year.
The zip file includes documentation and a sample
.BAT driver file/script.
Documentation
(in PDF format) - is also included in the above zip file.
2004-11-11. Sample data set for Micromass Qtof. See section Getting started for details.
2004-11-11.
MSQuantInstaller,2004-11-11,1.4a7.zip.
Installer for version 1.4a7.
(There is no official public release version yet).
Note: Please note that using current versions
of MSQuant for Micromass data it is
dependent on a component in the Finnigan XCalibur
software package for displaying spectra and thus only run
if XCalibur software package is installed.
Note: requires Microsoft .NET version 1.1.
Added:
Parse result (including quantitation result) can now
be saved and loaded.
Please note that for this to work the same set
of associated files must be selected when the
Open command is started.
This new feature can be used as a work-around
for memory related problems:
Let the program parse. Save. Restart program.
Open.
or:
Open. Quantitate a number of proteins. Save.
Restart program. Open. Quantitate more proteins. Etc.
Changes:
Only validated peptides are now MS3 scored.
Now more memory efficient, especially for Finnigan data.
Bug fixes: MS3 scoring now uses real peaks instead of
data points (important for spectra acquired in profile mode).
MS3 scoring now works for long peptides.
2004-08-09. Sample data set for LTQ-FT. See section Getting started for details.
2004-08-02.
Warning: We recently discovered that
uninstalling older versions of MSQuant (that you have
to do when using a new installer) causes the
QSTAR/Analyst installation to break, amongst other
things preventing display of QSTAR spectra in
Analyst QS.
This problem has been fixed in later
MSQuant installers. The easiest cure at this point
in time is to reinstall Analyst.
We are sorry for any inconvenience caused.
2004-07-23.
MSQuantInstaller,2004-07-23,1.4a6.zip.
Installer for version 1.4a6.
(There is no official public release version yet).
Note: in this version Micromass support has
been reenabled. However please note that it is
dependent on a component in the Finnigan XCalibur
software package for displaying spectra and thus only run
if XCalibur software package is installed.
Note: requires Microsoft .NET version 1.1.
Added:
Double labeling; 2 amino acids can now have
different masses, e.g. +6 Da for Arginine and
+4 for Lysine.
Changes:
in MiscOptions.xml there are now separate
values for quantitation XIC mass window, one
for each of the three instruments supported;
QSTAR, LTQ-FT and Qtof; the old field
"absoluteMassWindowForXIC" is now ignored.
Micromass support reenabled.
In the protein list window quantitation and MS3
scoring now works on the selection and the
selection can now contain more than one
selected item.
Better zooming the in the spectrum display in
the quantitation window.
An LC profile is now computed/displayed even if
the corresponding signal in another LC
profile is identical zero.
2004-07-21. MSQprojectFiles,2004-07-21.zip. For MSQuant developers only. Updated sample project files for users of Visual Studio .NET.
2004-07-19. MSQuantInstaller,2004-07-19a.zip (3.7 MB). This version includes Micromass support!! Installer for possibly unstable development version as of 2004-07-19. This version was mainly released to demonstrate Micromass support. Version 1.4a6 will soon be released.
2004-06-28. MSQuantInstaller,2004-06-28a.zip. Installer for version 1.4a5. Note: in this version Micromass support is turned off, but an installer with Micromass support turned on will soon be placed here. (There is no official public release version yet). Note: requires Microsoft .NET version 1.1. Changes: More than 100 LC points can now be plotted in the lower right corner of the quantitation window- no longer leaves out the right part of some LC profiles. Added: Spectrum markers for Finnigan spectra, both in protein validation window and in quantitation window. MS3 scoring for all ms3 spectra, not just the first; for scoring the spectrum that gives the highest score is used. Accepts yet another format of the Mascot result files (resulting from an update on our Mascot server computer). For protein ratio (quantitation) the program now uses intensity weighted peptide ratios instead of non-weighted peptide ratios. Spectrum markers in Finnigan spectra now also works when the user zooms. Intensities of identical zero are ignored when finding peaks in LC profiles (particular important for Finnigan data).
2004-06-07. MSQuantInstaller,2004-06-07a.zip. Installer for version 1.4a4. Note: in this version Micromass support is turned off. (There is no official public release version yet). Note: requires Microsoft .NET version 1.1. Changes: MS3 scoring. Fragment labels for Finnigan spectra. Return-key can now be used instead of double-click in: main window, Protein List window and Protein Validation window. Now asks for confirmation before closing the program and closing the protein list window.
2004-03-22. MSQuantInstaller,2004-03-22a.zip. Installer for version 1.4a3. Changes: Support of Mascot 2.0 HTML output format. Support of Micromass files.
2004-02-16. MSQprojectFiles,2004-02-16.zip (no longer available). For MSQuant developers only. Updated sample project files for users of Visual Studio .NET.
2004-01-29. MSQuantInstaller,2004-01-29a.zip (no longer available). Installer for version 1.4a2, latest development version. (There is no official public release version yet). Changes: the quantitation modes and recognised modifications are now defined externally to the program, in MSQ_quantitationModes.xml.
2004-01-27. MSQuantInstaller,2004-01-27a.zip (no longer available). Installer for version 1.4a1.
2004-01-27. MSQprojectFiles,2004-01-27.zip (no longer available). For MSQuant developers only. Sample project files for users of Visual Studio .NET.
1. Download and install Microsoft .NET (if not already installed). Version 2.0 or later of .NET is required for MSQuant versions 1.4a17 and later.
2. Install software from at least one mass spectrometer vendor. Currently Sciex/Analyst and/or Finnigan/XCalibur and/or Micromass/MassLynx is required/supported.
Notes for installation:
3. Download an MSQuant installer from our Web-server. (It is placed on our web-server because I can’t get the SourceForge file release system to work.)
4. Install MSQuant by running the installer (setup.exe). Note: you may get error messages for the mass spectrometer vendor software that you don’t have installed or need. E.g. if you only use Analyst/QSTAR you will get 2-3 error messages for files belonging to the Finnigan XCalibur package. You can safely ignore those error messages.
5. For QSTAR/Sciex/Analyst users only: if you have Analyst QS version 1.0 installed (in contrast to version 1.1) then you must do the following to use later versions of MSQuant:
1. For Analyst: the so-called system path must contain the Analyst bin folder otherwise you will get an error message about some DLL that can not be found (FMWIFFCompDocNTDriver.dll). What to do:
2. For Finnigan/XCalibur: during installation of
XCalibur "XDK" must have been checked (it is off by
default). If you have already installed it without
the XDK then you should take steps to remedy this
situation; e.g. reinstalling with XDK checked should
work. Disclaimer: do it at your own risk. Before you
reinstall save any changes you have made to
your existing installation, e.g. extra modules that
were added after initial installation.
MSQuant is very memory hungry, depending on the size of the search result and possible use of retention time cross-correlation. We recommend as much RAM in the computer as possible. We strongly recommend setting the virtual memory to the maximum of 2 GB. This can be done as follows: Right-click on "My Computer" on the desktop. Properties. tab Advanced. Performance Options. Change. Set "Initial size (MB)" to 1700. Set "Maximum size (MB)" to the same number, 1700. OK. OK. OK. It may be necessary to restart the computer.
1. Download one of the example raw file sets:
2. Open the program.
3. Set appropriate raw data file type (menu Tools/Options, Raw file type, QSTAR/LTQ-FT/Qtof.)
4. Set appropriate quantitation mode, matching the data (menu Tools/Options, Quantitation mode, Arg13C6/Lys13C6/Arg13C6 and Arg13C615N4/HysTag/No isotope/N15 labeling)
5. Prepare for parsing the Mascot result file:
6. Double-click on the Mascot HTML result file in the left pane. This will start parsing the protein and peptide information from the Mascot HTML result file. Eventually the Protein List window will be filled with the proteins from the Mascot search result file.
7. Double-click on the top protein. This will open the Protein Validation window.
8. Double-click on a peptide. The MS-MS spectrum will be displayed. For this to work and all other operations that requires access to the raw spectrum see the section below, "Preparing input files for the Mascot search engine".
9. Quantitate:
If all this works then you are ready to try it on your own data. The input file for the Mascot search must be formatted in the way that MSQuant expect. See next section, "Preparing input files for the Mascot search engine".
MSQuant uses the retention time (for the MS-MS event) for a peptide in order to find the MS-MS and corresponding MS spectrum in the raw spectrum file.
This information is expected by MSQuant to be in a particular format. It should be placed in the TITLE fields in the Mascot input file (in Mascot Generic Format, file extension .mgf, .msm, .tmp or .txt). Within the TITLE field it should be in the sub-field "Elution from:" Sample (for QSTAR):
TITLE=Elution from: 31.16 to 31.16 period: 0 experiment: 4 cycles: 1
Elution = retention time. Unit is minutes.
Version 1.1 of Analyst: Mascot.dll now creates a bastard output format. In order to use MSQuant a script, QSbroken2QSclassic, must be run to convert the output of Mascot.dll to an appropriate format. The script expects a file in the Mascot Generic Format (MS/MS Ions Search) as input, but does not require a particular file extension. The nominal file extension is .MGF, but different tools uses different file extensions. We know of 3 other file extensions: .TMP, .MSM and .TXT. .TMP is used by the some versions of the Mascot.dll script (QSTAR/Sciex/Analyst). The output file is also in Mascot Generic Format. The file name, including file extension, is specified by the user, in the BAT file; QSbroken2QSclassic.bat. Thus you must first edit this BAT file to fit your particular location and file names. Then double click on it to start the conversion. Perl must be installed on your computer in order for this to work. For Windows computers we can recommend ActivePerl, http://www.activestate.com/activeperl/?psbx=1. The exact version number does not matter.
Version 1.0 of Analyst: the built-in tool, Mascot.dll, or the proprietary tool "IDA processor" should create the correct format.
For Finnigan DTA files the retention time information is not provided. Instead the spectrum number is encoded in the file name. Example: "overnight.1169.1169.3.dta". This spectrum number must be put into the field "FinneganScanNumber", otherwise the retention time is read from the field "Elution from". Sample:
TITLE=Elution from: 777.777 to 777.777 period: 0 experiment: 1 cycles: 1 FinneganScanNumber: 1169
If the field "FinneganScanNumber" exists then MSQuant ignores fields "Elution from"/"to" - FinneganScanNumber overrides those fields.
A .NET application, DTASuperCharge, is available
to convert one or more Finnigan .RAW files to
Mascot search input files in a format suitable
for use with MSQuant.
DTASuperCharge is bundled with later versions of
MSQuant and this is currently the only way of
getting recent versions of DTASuperCharge.
Installation:
To create the input file for Mascot, some options need to be set correctly. In the ProteinLynx module (NOT ProteinLynx Global Server) in Masslynx, the option "Append data to single file" should be unchecked. In this case the MS/MS channel number, the scan number and the charge are encoded in all files. Example: "overnight.117.3.2.pkl". Copy the Perl script and the driver BAT file (see below) to the output directory, edit the BAT file as appropriate and execute it.
A Perl script, PKL2Mascot.pl, is available to prepare .PKL (peak list) files for search for Mascot such that MSQuant will be able to find the MS-MS and MS spectra. Location: PKL2Mascot,2004-02-21.zip.
The output of the script is the input to the Mascot search. Perl must be installed on your computer in order for this to work. For Windows computers we can recommend ActivePerl, http://www.activestate.com/activeperl/?psbx=1. The exact version number does not matter.
Part of the output of this script will look similar to:
BEGIN IONS
PEPMASS=446.7263
CHARGE=2+
TITLE=Elution from: 9999.99 to 9999.99 period: 0 experiment: 3 cycles: 1 FinneganScanNumber: 117
183.1330 11.0000
228.1346 13.0000
261.1865 14.0000
552.2947 11.0000
665.4302 8.0000
762.4420 13.0000
END IONS
Regarding .IDX files: the .RAW folder contains several .IDX files. MSQuant needs to know what the location of the .RAW directory is and this is done by selecting an .IDX file (does't matter which one).
MSQuant can handle multiple raw data files at a time, e.g. for protein correlation profiling (PCP). This is described in further detail in a separate document. It relies on a Perl script to put the necessary raw file name information into the Mascot search input file. All the files (scripts and documents) for this can be downloaded as a single ZIP package.
Note 1: this currently only works for QSTAR and LTQ-FT data, not Qtof (a future version of MSQuant may change this). Note 2: the script in this package has not been updated to understand Mascot.dll for QS 1.1 (or later) output. You can work around this by using the QSbroken2QSclassic script.
We have developed procedures and scripts for handling N15 labeling samples.
Updated on 2009-10-06 by Peter Mortensen (
Email address
).
Updated on 2009-10-02 by Peter Mortensen (
Email address
).
Updated on 2009-09-11 by Peter Mortensen (
Email address
).