CHANGELOG

26.11.2019 LipidXplorer 1.2.8.1

Bug Fixes
- Fixed broken Import Settings "Select a Configuration Save As..." and "Delete"
- Added additional log output to print time limits applied to scan selection in seconds and minutes

26.11.2019 LipidXplorer 1.2.8.1

Bug Fixes
- Fixed broken Import Settings "Select a Configuration Save As..." and "Delete"
- Added additional log output to print time limits applied to scan selection in seconds and minutes

21.09.2019 LipidXplorer 1.2.8

New Functionality
- Added MSFilter and MSMSFilter
The MSFilter and MSMSFilter takes a value between  0 and 1 can be chosen representing the fraction of scans in which a certain peak has to be present.
The filter is applied for both MS and MS/MS according to the given import setting for each scan type. 
For each precursor ion all associated MS/MS scans are collected according to selection window. Counting of MS/MS scans and association of precursor is done automatically. 
Accordingly, the number of MS/MS scan can vary for each precursor ion in the shotgun acquisition, as might occur when using a DDA strategy.

- Improved mzML file loading speed from O(n^2) to O(n).
This improves loading speed of mzML files significantly and brings them on par with mzXML files considering import of data.

- Added application and taskbar icons.

- (Developers) Anaconda package management.
LipidXplorer now comes with a dependency definition for an Anaconda environment for development under Windows. Please check the GitLab README.md file for further information.

- LipidXplorer executable distributions (convenience binaries).
LipidXplorer is now distributed as a Windows and Linux archive containing executables that come with all dependencies included. These can be downloaded from https://lifs.isas.de/lipidxplorer.

Bug Fixes
- Updated Wiki locations to https://lifs.isas.de/wiki
- Fixed MFQL editor dialog closing error on Windows 10 and Python 2.7.
- Fixed application not exiting properly due to non-terminated threads.
- Improved file handle closing by switching to auto closing behaviour with 'with'.
- Fixed output file creation for user accounts with '.' characters in path.
- Fixed UI label truncations and spacing issues on Windows 10.
- Updated wxPython dependency from 3.0 to 4.0.4.

Removed Functionality*
- Removed sample grouping
- Removed DTA/CSV + PIS support for input spectra

Deprecated Functionality
- The mzXML format is deprecated and will be removed for the LipidXplorer 1.3.0 release.

* If you rely on this functionality, please contact us at lifs-support@isas.de

12.08.2008

- new function isStandard(<variable>, "<sample>", "<scope>"). Place it in SUCHTHAT section. 
  A scan is started to find <variable> in <sample>. Once found it is declared as the standard and the
  intensities of all other peaks in <scope> are recalculated as ratio. 
  Example: isStandard(varStandard, "hilde01", "MS1+")

- new function for addressing groups of samples with the help of placeholders:

    Patterns are Unix shell style:
    
    *       matches everything
    ?       matches any single character
    [seq]   matches any character in seq
    [!seq]  matches any char not in seq

  Example: FA1.intensity["*hilde0[1-9]*"] > FA1.intensity["*hilde1[0-9]*"]

- also possible: column(FA1, "*hilde0[1-9]*")

13.08.2008

- semi-new function: isStandard(<variable>, "<scope>"). With this syntax the standard is
  calculated for every single sample.

14.08.2008

- every peak (either MS or MS/MS) has now additionally the following information:
    . peak mean
    . peak median
    . peak variance
    . peak standard deviation
  all this info will be put out in the dump

22.08.2008

- isStandard() function works now for MS1+/- (! no MS2+/- !). The function should be placed in the SUCHTHAT
  section and has as attributes 1) a variable containing the marked standard and 2) the scope of
  the standard (MS1+, MS1-, MS2+ or MS2-). Next an example of an MFQL script identifying
  PC with calculating the standard:

	QUERYNAME = Phosphatidylcholine;
	DEFINE PR = 'C[36..50] H[30..200] N[1] O[8] P[1]' WITH DBR = (1.5,7.5), CHG = 1;
	DEFINE DietherPC = 'C44 H93 O6 N1 P1' WITH CHG = 1;
	DEFINE DietherPE = 'C45 H95 O6 N1 P1' WITH CHG = 1;

	IDENTIFY Phosphatidylcholine WHERE
		
		# marking
		PR IN MS1+ WITH TOLERANCE = 4 ppm OR
		DietherPC IN MS1+
	  
	SUCHTHAT
		isEven(PR.chemsc[C]) AND
		isStandard(DietherPC, "MS1+")
	  
	REPORT 
		MASS = "%4.4f" % "(PR.mass)";
		NAME = "PC [%d:%d]" % "((PR.chemsc)[C] - 8, (PR.chemsc)[db] - 1.5)";
		PRECURINTENS = PR.intensity;;

2.09.2008 

- de-isotoping was implemented for MS mode. The algorithm uses the sum compositions
  which where calculated with the used MFQL scripts. So, only molecular species of interest
  are considered for the de-isotoping. The algorithm is the following:
  1) sort MS spectrum increasingly.
  2) for every mass m which has a chemical sum composition assigned:
    3) look, if there is are masses i1, i2, i3 or i4 which could be isotopes, i.e.
      is there a mass i1 = m + 1.0033,
      is there a mass i2 = m + 2 * (1.0033), and so on ...
    4) calcuate the isotopic distribution of m for only 13C. Isotopes of other elements
       are so little for lipids, that we leave them beside. The distribution is binomial
       with a probability that a 13C occurs of 0.01082 <insert citation here>.
    5) substract calculated isotopic percentage from i1-4

- de-isotoping of MS/MS was revisited. The algorithm is the following:
  1) generate artificial PIS spectra P1, ..., Pn for fragments f1, ..., fn by collecting 
     precursor masses which have f1 (f2, ..., fn) in thier MS/MS spectrum.
  2) for every PIS spectrum Pi:
    3) for every mass m in Pi:
      4) look, if there is are masses i1, i2, i3 or i4 which could be isotopes
      5) calculate the isotopica distribution of the neutral loss of m accoding to
        <insert citation here>
      6) subtract the calculated isotopic percentage from i1-4 

1.10.2008

- implemented a debugging window in the GUI.

- GUI looks a bit more compact now

- update of the merging algorithm (for *.mzXML import). Average masses are now calculated 
  by intensity weighted average. This yields more accurate spectra.

9.10.2008

- switched off the cleaning procedure for *.mzXML imported files

- if more than one sum composition is found for a precursor mass, it will be reported
  ordered by identification error.

- Bug with *.mzXML file, where only MS/MS spectra are given is fixed.

23.10.2008

- new Function: complementMasterScan. Switch it on with a checkbox on the
  Run-Panel. It will produce the "complement MasterScan" of the current
  query, i.e. a MasterScan with all peaks which where not identified in
  the current run. It will be saved as <original MasterScan name>-complement.sc.

- the purpose is do blind queries for unknown or not excpected sum compositions.

18.11.2008

- new Function: DEFINE allows now to define a list of same variables with
  different names in one line. The user just writes a list:
    old:
      DEFINE FA1 = 'C[14..22] H[20..60] O[2] N[1]' WITH DBR = (0.0,6.0), CHG = 0;
      DEFINE FA2 = 'C[14..22] H[20..60] O[2] N[1]' WITH DBR = (0.0,6.0), CHG = 0;
      DEFINE FA3 = 'C[14..22] H[20..60] O[2] N[1]' WITH DBR = (0.0,6.0), CHG = 0;
    new:
      DEFINE (FA1, FA2, FA3) = 'C[14..22] H[20..60] O[2] N[1]' WITH DBR = (0.0,6.0), CHG = 0;
    
- new Function: DEFINE allow the definition of a list of fragments. For example:
    DEFINE FA = ('C14 H27 O2 N1',
                 'C14 H29 O2 N1',
                 'C14 H31 O2 N1',
                 'C15 H21 O2 N1',
                 'C15 H31 O2 N1');

2009-03-06

- adding a new Variable to TypeMFQL: currVars, which carries the current variables
  and is accessible for all functions
- adding a new function: avg(), which calculates the average of the attribute (which
  should be an intensity vector by now)

2009-07-02

- new function in the import which allows grouping the samples by letting the user put
  them into group folders. This does only work for *.mzXML until now.

2009-12-01

* fixed massrange error when using masses in DEFINE
* fixed problem with order of the *.mfql scripts in the Run Panel

2009-12-03

* implemented multiple charge recognition !NOTE!: it is to be tested!
  Lipids with charge unequal to 1, 0 or -1 are not able to account to
  isotopic correction.
* upgrade of the Exception handling.