Home
News
Docu
Publications
Team
Links

News

Change log (August, 23rd, 2007):

• DecoyDatabaseBuilder: Due to some changes in the format of the protein header for the UniProtKB protein database, it was necessary to change the identifier for decoy proteins:
  • shuffled decoy proteins accessions begin with 'sh'
  • reversed decoy proteins accessions begin with 're'
  • random decoy proteins accessions begin with 'ra'
  • the source proteins accessions stay the same, i.e. begin with a capital letter and a digit.

Change log (August, 20th, 2007):

• General: A new file format is supported. Peaklist files exported from the ProteinScape^TM software in MGF-format can be imported into Peakardt if the filename is changed to .mgf_ps. This is necessary, because the exported files do not comply to the MGF standard.

Change log (August, 9th, 2007):

• ExperiAna, Properties: Added some more statistical evaluations on multiple quantitative experiments and cleaned up the resulting output. There are now two output formats: simple and comprehensive. Both are selectable via radio buttons in the FindPairs tab in the Properties frame. Here are the values calculated from all the peptides from multiple experiments:
  • simple results table:
    • overall pairs found
    • protein ratio
    • p-value
    • deviation
  • comprehensive results table:
    • protein ratio (estimation of location):
      • median
      • arithmetic mean
      • lower and upper bound of the confidence interval
      • p-value
  • deviation (estimation of scale):
    • standard deviation
    • relative standard deviation (%)
    • Q_n and S_n estimator
    • deviation (%)
    • overall variance
    • biological variance
    • technical variance

Change log (July, 4th, 2007):

• DecoyDatabaseBuilder: fixed an parsing error that occurred when generating composite decoy databases from Uniprot/Swissprot databases. This was caused by unexpected symbols in the target databases protein header.

Change log (June, 14th, 2007):

• FindPairs: the automatic submission to the ExperiAna module has been deactivated, because sometimes the manual editing of the resulting pairs / peptides list is desirable. For sending the resulting peptides list to ExperiAna in the Experiment menu the menu item send result to ExperiAna has to be selected.

Change log (May, 22nd, 2007):

• Website: added some general information about the supported MS data file formats. and a tool to extract MS data from WIFF files for QSTAR and other instruments using the Analyst software from Applied Biosystems.

Change log (May, 18th, 2007):

• General: a new usermanager was implemented for the Peakardt software suite. For the user there should not be any changes visible.
• Website: some more explanations for protein quantitation and the usage of the FindPairs module is presented.

Change log (May, 9th, 2007):

• ExperiAna, ProteinAssembly: the usage of external libraries now work for users over the internet. The libraries do not have to be installed on the local computer and are loaded automatically over the internet if they are needed for specific functions.

Change log (April, 25th, 2007):

• ExperiAna: in order to clarify the resulting terms the fraction of variance for experiments was renamed to biological variance and the fraction of variance for peptides was renamed to technical variance.
• ExperiAna: if the p-value is very close to 0 (i.e. < 1E-308) the term "> 0" is displayed instead. Previously the value 0E0 was displayed which was due to a rounding error. The p-value is always different from 0 or 1.

Change log (March, 14th, 2007):

• General: The export of spectra to MGF was implemented. Now all importable file formats can be exported to DTA-directories, MGF- and CPL-files.

Change log (March, 12th, 2007):

• ExperiAna: The ExperiAna module for an integrated meta analysis of multiple experiments or repetitions was complemented. Now the statistical analysis contains:
  • arithmetic mean
  • median
  • deviation (in percent of the arithmetic mean), i.e. coefficient of variation
  • variance
  • fraction of variance for experiments (biological variance)
  • fraction of variance for peptides (technical variance)
  • lower and upper bound of significant peptides
  • p-value
  • arithmetic mean of the deviation of all experminents

Change log (February, 12th, 2007):

• FindPairs: while the FindPairs-algorithm is working on the spectra a small window informs about the relative progress reference to the number of spectra.
• Properties / Spectra and table display: when the setting for the displayed number format (e.g. number of fractions or use of number grouping) changes and the Apply or OK button is pressed in the Properties window, the displayed spectra and tables are automatically redrawn with the new parameters.

Change log (February, 8th, 2007):

• Spectra display / Properties: in the Properties window under the Display tab there is a new combo box for entering the font size of the scale which is displayed for the spectra. The default value is 12, but can be changed either by selecting a value from the list, i.e. 6 to 36, or by entering an arbitray other number from 1 to 127. After pressing the OK or Apply button in the Properties window the scales of spectra are immediately drawn with the new font size.

Change log (January, 29rd, 2007):

• ExperiAna: Because single quantitative experiments reveal questionable results, a new module for Peakardt was developed. Here you can manage your quantitative experiments and performed sound statistical analyses on your data. Enter quantitative peptide or protein ratios from multiple independent experiments and calculate a single ratio as a result. Calculation of arithmetic mean, median, standard deviation, T-Test and variance analysis are automatically performed.
• FindPairs: Added to the Start-menu a new item: Send result to ExperiAna. Changed the name of the Start-menu to Experiment-Menu to point out its new focus. The quantitative result of a FindPairs search can be send via this button to the ExperiAna module to be analysed in a row of experiments. The name of the button FindPairs was changed to start FindPairs.
• General: a new import format was implemented. Spectra from the HCT plus (Bruker Daltonik) mass spectrometer can be exported in an ascii format. This is similar to DTA format with a different header and contained charge states for every peak. Here is an example of the format. If the charge state is undefined and missing in the table, then the imported data contains two peaks with the same m/z and intensity but with a charge state of two or three.

Change log (January, 23rd, 2007):

• website: somewhat clarified the description of the Decoy Database Builder.

Change log (January, 9th, 2007):

• DatabaseSearch: the found proteins are no longer displayed as a list, but as real sortable table. The table is now integrated in the database search window. A preliminary score for the peptide search is shown in a new column besides the protein accession number, sequence length and number of peptide hits. Besides that, the protein database search performance was considerably improved: the search for a short peptide sequence in the uniprot database revealed in 3,5 sec that 1248 proteins contained that sequence. The time includes the search and display of the results.
• ProteinAssembly: the parsing of the input peptide sequences has been improved and the theoretical peptide mass is automatically calculated. As a target for calculating the original protein sequence from the peptide sequences the previously measured original protein mass can be entered.

Change log (January, 2nd, 2007):

• DecoyDatabaseBuilder: a command line version of the DecoyDBB has been implemented. If you want to run the software in a larger bioinformatics infrastructure then the DecoyDBB can now be called from within batch scripts. The syntax for generating decoy databases is as follows:
  

  java -jar DecoyDBB filename decoytype append masstolerance

  • java: the path to the java interpreter (java.exe)
  • -jar: command line option of the java interpreter to specify that the following java program resides in a java archive (JAR)
  • DecoyDBB: the name of the Decoy Database Builder JAR file (without the JAR extension). Make sure that the DecoyDBB.jar file is in the current directory.
  • filename: the complete path to the protein database file for which a decoy DB should be generated.
  • decoytype: a number to specify the type of the decoy database:
    • 0: reverse
    • 1: shuffle
    • 2: random
    • 3: combine (reverse+shuffle+random)
  • append: should a hybrid decoy database be generated?
    • true: the source and the decoy proteins are contained in the decoy database
    • false: only the decoy proteins are contained
  • masstolerance: for random (and combined) databases a mass tolerance has to be specified

Change log (December, 15th, 2006):

• Website: when visiting the Peakardt homepage and the Java plugin of the browser was not correctly installed, then the automatic display of the JRE download link was incorrect. This was corrected.

Change log (November, 29th, 2006):

• DecoyDBB: the header format of the UniProtKB / SwissPort database was changed, so an adjustment of the header processing for decoy databases was performed.

Change log (October, 25th, 2006):

• mzXML: support for mzXML spectra files was completed. Now also PFF spectra can be imported.

Change log (October, 18th, 2006):

• mzXML: added fundamental support for the mzXML spectra data format. If a mzXML file is opened via File | Open PMF... the xml file is parsed for all precursor ions with included m/z values. The total ion current represents the intensity value. The contained charge state value in the mzXML file is used, but set to -1 if not present.

Change log (October, 12th, 2006):

• 2D Spectra: the solution of the 2D Spectra is now depending on the Mass Tolerance set in the Properties | Filter tab. Peaks within the mass tolerance are merged, their intensities are added. If the Mass Tolerance is set to e.g. 1 Da then the 2D Spectra show one pixel per Dalton. Setting the Mass Tolerance to e.g. 0.1 increases the size of the m/z axis by ten.
• General: If an MGF-fragment spectra file is opened via "Open PMF..." in the File menu, then the file is scanned for peptide parent masses. The found masses and appropriate charge states are used to build a reconstructed peptide mass fingerprint spectrum. By default the charge state of each peptide is set to 1.

Change log (September, 25th, 2006):

• FindPairs: after some discussion among computer scientists and statisticians, the calculation of the coefficient of variation has slightly changed. The old coefficient was not wrong, but calculated 'only' the sample variation and was so seen controversial among statisticians. The new version calculates the population variation. The quotient in the variation formula changed from N to (N-1). If the sample only contains one element, the coefficient is set to 0.
  For some more information take a look here.

Change log (September, 20th, 2006):

• SetOperations: the Set Operations software was put on the web server to be publicly available. If you want to use it, please send me an email and you will receive your individual password. You also have to enter your IP address from which the software will be used.

Change log (September, 19th, 2006):

• DecoyDatabaseBuilder: the DecoyDBB is password protected from now on to avoid not authorized usage. If you want to use it please send me a short email and you will receive a reply with your individual password. Please specify the IP address from which you will use the DecoyDBB. If you want to use the DecoyDBB in the future the password and IP address have to match.

Change log (August, 23rd, 2006):

• DecoyDatabaseBuilder: the Uniprot / Swiss-Prot (UniProtKB) protein databases can now be processed by the Decoy Database Builder.
  • decoy accession identifiers:
• shuffled decoy proteins accessions begin with 'S'
• reversed decoy proteins accessions begin with 'E'
• random decoy proteins accessions begin with 'A'
• the original proteins accessions stay the same, i.e. begin with 'O', 'P' or 'Q'.
• Example:
  • >108_LYCES (Q43495) Protein 108 precursor
    MASVKSSSSSSSSSFISLLLLILLVIVLQSQVIECQPQQSCTASLTGLNVCAPFLVPGSP
    TASTECCNAVQSINHDCMCNTMRIAAQIPAQCNLPPLSCSAN
    the randomized sequence looks like this:
  • >108_LYCES (A43495) Protein 108 precursor
    CSVLSKSNTLVTMMRSVPIASSAEPPCQANSALGISARVCASTLGIISQVQPSTVNASSAPSFINLCILNNFSTATVDMF LMLMSLATSPNQSMICNSTISQ

Change log (August, 21st, 2006):

• DecoyDatabaseBuilder: if database statistics for randomizing decoy sequences are used now no longer only fixed statistics are used, but the statistics of the protein database at hand. Therefore the amino acids of the database has to be counted to determine the overall statistics. This works for fasta files as well as pasted sequences in the input text field.

Change log (August, 18th, 2006):

• Website: the website has undergone some changes. Besides some minor layout changes of the menu, the stand-alone versions of the Decoy Database Builder and Set Operations were made available from the main page. The Decoy DBB is available for everyone, while the Set Operations application and the whole Peakardt suite still have restricted access. If you are interested in participating in software test, please send me a note.

Change log (August, 17th, 2006):

• Peakardt: in the Help menu direct links to some websites were added:
  • Peakardt.org
  • internal homepage
  • Medizinisches-Proteom-Center.de

  if those menu items are selected, the default web browser is started and the desired page is opened therein

Change log (August, 16th, 2006):

• DecoyDatabaseBuilder, Properties: added a new parameter to control the generating of random protein sequences. Now you can choose between the probabilities of amino acid occurrence in this particular protein or the whole database. The desired option can be selected in Properties | EvAl heuristics: use protein statistics or use DB statistics. For the database approach the following numbers of amino acid occurrences are used (from NCBInr version of 23.03.2006):
  • A: 98082879
  • C: 17764055
  • D: 62304280
  • E: 73927235
  • F: 47806446
  • G: 83403687
  • H: 27220335
  • I: 68691642
  • K: 64098270
  • L: 117231914
  • M: 28157708
  • N: 50626598
  • P: 59685956
  • Q: 48543919
  • R: 66066561
  • S: 84756302
  • T: 66821484
  • U: 640
  • V: 78248027
  • W: 15691039
  • Y: 36061470

Change log (August, 14th, 2006):

• DecoyDatabaseBuilder, EvAl Peptide Identifier : Fixed some errors that occurred while generating random peptide sequences. The error had a probability of only 1:2³¹ to appear, so it did not really made big impact on decoy databases.

Change log (August, 9th, 2006):

• Digest: The protein and peptide counter displaying the progress of the digestion now correctly logs the advancement. For the first thousand proteins the overall peptide sum is displayed. After that the display is updated only every thousand proteins.

Change log (August, 5th, 2006):

• FindPairs: Search result files can be opened for importing identified peptide sequences. Currently only Sequest's OUT-files are supported. Select from the File menu Open Search Result... and chose all OUT-files to be imported. According to the parameters specified in Properties | Assembly, i.e. minimum value for Delta Cn and XCorr as well as the maximum rank, the appropriate peptide sequences are entered into the sequence field in the Pairs tab of the FindPairs frame.

Change log (August, 4th, 2006):

• SetOperations, FindPairs, Digest: optimized memory usage for saving files in these modules. This results in far less memory overhead and higher speed for saving files to disk in some methods. Moreover larger result files can now be saved.

Change log (August, 4th, 2006):

• SetOperations: added a new feature: count all elements of a list. A list of elements, e.g. peptide or protein sequences/accession number, may be entered by copy and paste, or by the appropriate menu item. After pressing the start counting button for each unique element in the list its absolute appearance is counted and displayed in a sorted table.

Change log (August, 3rd, 2006):

• DatabaseSearch: improved the building of Suffix Arrays. Besides some minor internal improvements, now the whole accession number is displayed for each matching protein. This works for IPI, NCBI, Swiss-Prot, Trembl and other protein databases.

Change log (August, 2nd, 2006):

• PeptideAssembly, DatabaseSearch: Search result files can be opened to retrieve the identified peptides. Currently Sequest's OUT files are supported. The number of peptides can be limited by choosing specific parameters in the Properties, so not all peptides are imported but only those matching the criteria.
  • considered Sequest parameters are:
    • minimum delta Cn. The peptide should have at least this delta Cn value. Default value is 0.0.
    • minimum XCorr. The peptide should have at least this XCorr value. Default value is 0.0.
    • maximum Rank. The peptide should not have a higher rank than specified. Default value is 2.

Change log (July, 13th, 2006):

• OpenPFF, OpenPFFaio, OpenPMF: a warning message is now displayed if an ill formatted spectra file is opened.
• Digestion, Fragmentation, FindPairs: added a new modification display style for sequences. If in Properties|Modification the option diamond is selected via the appropriate radio button, then instead of the name or composition of the modification only a small diamond (◊) is displayed behind the modified site.

Change log (July, 11th, 2006):

• FindPairs: Fixed a bug in FindPairs that appeared when the digestion of a protein sequence lead to a modified peptide pair where the first modified peptide has a higher mass than than its counterpart.
• FindPairs: added a new menu to the frame as a shortcut to run the FindPairs algorithm: Start|FindPairs.

Change log (June, 27th, 2006):

• SetOperations: if large input lists (several million entries) are used for SetOperations, the memory resources for the Java heap space may run short. For that purpose the display of the input and output lists may be omitted to save memory. If the new show lists checkbox is selected and input files were opened via the File menu then those input lists are not displayed. The results lists are likewise not displayed, but the number of resulting set members is shown on top of the text areas.

• Garbage Collection: added a new tool to the Tools-menu. If the garbage collection is started the software tries to allocate some more memory space for later use. This is reasonable when dealing with large data sets e.g. a lot of spectra files or big databases for digestion.
• Modifications: the modification definitions have been updated.
  • 200 new modifications were added:

  (3-aminopropyl)(L-aspartyl-1-amino)phosphoryl-5-adenosine, 10-ethoxyphosphinyl-N-(biotinamidopentyl)decanamide, 2,5-dimethypyrrole, 3,4-didehydroretinylidene, 3-(carbamidomethylthio)propanoyl, 3-methyl-2-pyridyl isocyanate, 4,4,5,5-D4 Lysine, 4-methyl-delta-1-pyrroline-5-carboxyl, 4-sulfophenyl isothiocyanate, 4-sulfophenyl isothiocyanate (Heavy ¹³C), 4-trimethyllammoniumbutyryl-, 5-hydro-5-methylimidazol-4-one, 5-hydroxy-N6,N6,N6-trimethyl, 6-N-biotinylaminohexyl isopropyl phosphate, Acetaldehyde +26, Acetaldehyde +28, ADP Ribose addition, alpha-amino adipic acid, amidino, Aminoethylbenzenesulfonylation, aminoethylcysteine, AMP binding site, Arginine oxidation to gamma-glutamyl semialdehyde, Biantennary, Biantennary (-1 galactose), Biantennary (-2 galactose), Biotin polyethyleneoxide amine, biotinoyl-iodoacetyl-ethylenediamine, bis-N-I-sulfonerahodamine, Bisphenol A diglycidyl ether derivative, Both ends of crosslink attached to same peptide, bromination, Bruker Daltonics SERVA-ICPL(TM) quantification chemistry, heavy form, Bruker Daltonics SERVA-ICPL(TM) quantification chemistry, light form, C terminal -K from HC of MAb, C-term blocking imide, ¹³C and ¹⁵N label, ¹³C label, C8-QAT, CAMthiopropanoyl of Lys, carboxyethyl, Carboxylation, cholesterol ester, cis-14-hydroxy-10,13-dioxo-7-heptadecenoic ester, copper sulfido molybdopterin cytosine dinuncleotide, covalent modification of lysine by cross-linking reagent, Cross-link of (Iodo)-uracil MP with W,F,Y, Crotonaldehyde, Cy3 CyDye DIGE Fluor saturation dye, Cy5 CyDye DIGE Fluor saturation dye, cyano, Cysteine mercaptoethanol, Cysteine modified Coenzyme A, cysteine oxidation to sulfinic acid, Cysteine sulfenic acid, Cysteinylation, D-Succinimide, d5-phenyl isocyanate, d9-4-trimethyllammoniumbutyryl-, Deamidation in presence of O18, deglycosylated asparagine, Dehydro, Dehydroalanine (from Cysteine), Dehydroalanine (from Tyrosine), Dehydropyrrolizidine alkaloid (dehydroretronecine) on cysteines, deuterated methyl ester, Deuterium Methylation of Lysine, diacylglycerol, didehydro, Diethylation, analogous to Dimethylation, diglutamyl, dihydroxy, Diisopropylphosphate, dimethyl 3,3'-dithiobispropionimidate, DiMethyl-C13HD2, DiMethyl-CHD2, dimethylated arginine, Diphthamide, dipyrrolylmethanemethyl, Ethanolation of Cys, Ethylation, flavin mononucleotide, fluorescein-5-maleimide, fluorescein-5-thiosemicarbazide, Fucose, Fucosylated biantennary, Fucosylated biantennary (-1 galactose), Fucosylated biantennary (-2 galactose), glucosylgalactosyl hydroxylysine, glucuronosyl, glycerophospho, glycerylphosphorylethanolamine, glycosyl-L-hydroxyproline, glycosylphosphatidylinositol, Half of a disulfide bridge, Heavy Sulfanilic Acid (SA) C13, heme, Heptose, Hexosamine, hydrogenase diiron subcluster, hydroxycinnamyl, hydroxyfarnesyl, hydroxyheme, Hydroxylethanone, hydroxymethyl, hypusine, IDBEST tag for quantitation, L-selenocysteinyl molybdenum bis(molybdopterin guanine dinucleotide), lactic acid from N-term Ser, Lactosylation, Levuglandinyl - arginine hydroxylactam adduct, Levuglandinyl - arginine lactam adduct, Levuglandinyl - lysine hydroxylactam adduct, Levuglandinyl - lysine lactam adduct, Light Sulfanilic Acid (SA) C12, lipid, MDA adduct +54, MDA adduct +62, Menadione derivative, Mercury Mercaptan, Methyl methanethiosulfonate, Michael addition of hydroxymethylvinyl ketone to cysteine, Michael addition of t-butyl hydroxylated BHT (BHTOH) to C, H or K, Michael addition with methylamine, microcin E492 siderophore ester from serine, molybdenum bis(molybdopterin guanine dinucleotide), molybdopterin, Monobromobimane derivative, monoglutamyl, monoisopropyl phosphate, monomethylated arginine, N-acetylglucosamine-1-phosphoryl, N-carboxylation of Met, N-ethylation, N-methylation, N-palmitoyl, N-pyruvic acid 2-iminyl, N-reductive amination-D, N-Succinimide, naphthalene-2,3-dicarboxaldehyde, Nethylmaleimidehydrolysis, Nmethylmaleimide, Nmethylmaleimidehydrolysis, nucleophilic addition to cytopiloyne+H2O, nucleophilic addtion to cytopiloyne, O-Sulfonation, O18 Labeling of Serine, Threonine or Tyrosine, O3-(riboflavin phosphoryl), octanoyl, One end of crosslink attached, one end free, Ornithine from Arginine, Oxidation with neutral loss of 64 Da, oxidized Lysine biotinylated with biotin hydrazide, oxidized Threonine biotinylated with biotin hydrazide, oxoalanine, oxobutanoic acid from N term Thr, palmitoleyl, persulfide, phenyl isocyanate, phenyllactyl from N-term Phe, phospho-guanosine, phosphoglycosyl-D-mannose-1-phosphoryl, phosphoribosyl dephospho-coenzyme A, phosphorylation to pyridyl thiol, phycocyanobilin, phycoerythrobilin, phytochromobilin, Pierce EZ-Link Biotin-HPDP, proline oxidation to pyroglutamic acid, Propionaldehyde +40, propyl-1,2-dideoxy-2'-methyl-alpha-D-glucopyranoso-[2,1-d]-Delta2'-thiazoline, pyridylacetyl, Pyrrolidone from Proline, pyruvic acid from N-term cys, pyruvic acid from N-term ser, quinone, reduced 4-Hydroxynonenal, reduction, retinal, S-(4a-FMN), S-diphytanylglycerol diether, S-Ethylcystine from Serine, S-nitrosylation, Selenium replaces sulphur in cysteine, selenyl, succinylated N-term Trp, Sulfitolysis, sulfonation of N-terminus, tetraglutamyl, tetraiodo, thiocarboxylic acid, thiophosphate labeled with biotin-HPDP, Thiophosphorylation, Threonine to a-aminobutyrate, Trideuteration, triglutamyl, trihydroxyphenylalanine, triiodo, Tryptophan oxidation to oxolactone, uridine phosphodiester

  • In total 392 modifications are now contained in Peakardt.
• Modifications: Unimod changed the XML-scheme for their modification database. The web-update mechanism in Peakardt has been adapted and is now able again to receive new modification definitions.

• FindPairs/Properties: to consider the consistency of an isotopic pattern of a peptide, select the consider intensities checkbox and enter an m/z value in the neighboring textfield. If the FindPairs algorithm finds an isotopic pattern, the height of monoisotopic peak is compared to the other peaks. If its mass is lower than indicated in the textfield and the intensity is higher, then the monoisotopic peak is considered to be successfully determined. If the mass is higher than the entered value then the intensity of the monoisotopic peak should be lower to be considered as part of a consistent isotopic pattern. This is an additional quality criterion to consider only non overlapping peptide pattern for quantitation.

• 2D Spectra: removed the intensity legend at the bottom of each 2D spectra image, because it was obsolete. The ColorScale shows the precise intensity information of each color value.
• 2D Spectra: The background color of the 2D spectra images has been changed from black to white, in order to give it a more familiar look according to 2D gel images. The assignment of color values to intensity values remained the same; take a look at the color scale.
  • here is an 'old' 2D spectrum with a black background.
  • here is the 'new' 2D spectrum with a white background.

• Receive Sequences: Protein sequences with from the UniProt knowledge base are also retrievable besides proteins from the IPI and NCBI databases.
• Modification: Additionally to lysine (K) the ICPL light and ICPL heavy modification may be attached to the N-terminus.

• Open PMF: The exported files of the Analyst software sometimes contain "Undefined" peak area values. This is most presumably occurring in low intensity peaks. When importing such peaks previously the software stopped importing the whole spectrum and issued a parse error. Now only the peak with the undefined peak area is omitted. The remaining peaks of the spectrum are imported. An error message is displayed with the mass of the omitted peak and the advice to use intensity values instead of peak areas.
• Randomizer, Decoy Database Builder, EvAl: Adapted the internal Randomizer to also be used with selenocysteine. The NCBInr database of 09/05/2006 contained 647 selenocysteins.
• Enzymes: Added selenocysteine as an unspecific cleavage site. Reworked layout and appropriate algorithms.

• Properties: If an improperly formatted value is entered in a text field where a numerical value should be entered, now a message box appears that a wrong parameter has been specified.
• Find Pairs: Changed the protein table HTML file format to match the displayed protein table. The peptide number now also starts with 1 instead of 0. If the peptides of the found pairs should have the same charge state then only the charge state of the 'light' peptide is saved, as it is the same than the charge state of the 'heavy' peptide.

• Open PMF: If exported peaklists from the Analyst software, ending with _qstar.txt, contain very low intensity peaks, e.g. < 0.001, the intensities are written in scientific notation. Because the value contains a small e instead of a capital E this lead to a wrong parsing of the value. Only the value before the e was used as intensity. This issue was corrected for intensity and peak area values, and the peak has now the proper value.

• Retrieve Sequences: a new tool was implemented. Upon entering the accession number of an arbitrary IPI or NCBI database the FASTA formatted protein header information and the amino acids sequence is displayed. The resulting sequences can be saved as a new protein DB on the hard disk.

• Find Pairs: the peptide intensity ratio in the protein table is now calculated by dividing the intensity sum of the 'light' peak cluster by the intensity sum of the 'heavy' peak cluster for each peptide.

• Modifications: added values for the ICPL labeling method in order to use them with FindPairs:

  • Code Name	Full Name	Mono mass	Avg. Mass	Sites	Composition
    ICPL light	Isotope Coded Protein Label light	105.0215	105.0215	K	C(6) N O
    ICPL heavy	Isotope Coded Protein Label heavy	111.0419	111.0419	K	C13(6) N O

• spectra import: optimized and broadened the MGF-file import. The Mascot Generic File format is supported by many MS instrument vendors to store spectra information but is sometimes non compliant to the Mascot standard. Peakardt is now able to read MGF files generated by at least ABI QSTAR, ABI Q TRAP, Thermo LTQ and the original format by Matrix Science.

• Decoy Database Builder: to be consistent with the the naming scheme of the NCBI and to avoid possible shortening of IPI accessions in the far future, the naming scheme of the decoy proteins has been changed. Instead of changing the accession numbers themselves, now the database identifiers are changed.
  • IPI decoy accession identifiers:
    • shuffled decoy proteins accessions begin with 'SHD'
    • reversed decoy proteins accessions begin with 'RED'
    • random decoy proteins accessions begin with 'RAD'
    • the original proteins accessions stay the same, i.e. begin with 'IPI'
  • NCBI decoy accession identifiers:
    • shuffled decoy proteins accessions begin with 'sh'
    • reversed decoy proteins accessions begin with 're'
    • random decoy proteins accessions begin with 'ra'
    • the original proteins accessions stay the same, i.e. begin with 'gi'
  • examples:
    • NCBI example shuffled:
      • >gi|385869|gb|AAB26507.1| type XI collagen alpha 1 chain [cattle, vitreous humor, Peptide Partial, 26 aa, segment 7 of 7]
        GLPGTQGSPGAGDGGIPGPAGPIGPP
      • >sh|385869|gb|AAB26507.1| type XI collagen alpha 1 chain [cattle, vitreous humor, Peptide Partial, 26 aa, segment 7 of 7]
        PQGDPPPAGPGLSGAGIITGGGPGPG
    • IPI example random:
      • >IPI:IPI00000001.1|SWISS-PROT:O95793-1 Tax_Id=9606 Splice Isoform Long of Double-stranded RNA-binding protein Staufen homolog 1
        MSQVQVQVQNPSAALSGSQILNKNQSLLSQPLMSIPSTTSSLPSENAGRPIQNSALPSAS...
      • >RAD:IPI00000001.1|SWISS-PROT:O95793-1 Tax_Id=9606 Splice Isoform Long of Double-stranded RNA-binding protein Staufen homolog 1
        QSQLNSQLMAQPGAQQSKNPKLSNGLQNVLPRQSPPSNVQGNPPRSSQVQSNRPWW...

• Decoy Database Builder: the naming scheme for the decoy accession identifiers was adapted for use with NCBI databases. Because of a different accessions numbering scheme, the decoy identifiers of the IPI do not work. In the NCBInr of 3rd March already 13968 protein entries exist beginning with '99'.
  • NCBI decoy accession identifiers:
    • shuffled decoy proteins accession number begin with 'sh'
    • reversed decoy proteins accession number begin with 're'
    • random decoy proteins accession number begin with 'ra'
    • the original proteins accession number stay the same, i.e. begin with a digit
  • the decoy protein accession identifiers are added in front of the original accession number. This results in an decoy accession number which contains two more characters.

• Decoy Database Builder: reworked the identification of decoy proteins within the decoy databases:
  • IPI decoy accession identifiers:
    • shuffled decoy proteins accession number begin with '99'
    • reversed decoy proteins accession number begin with '88'
    • random decoy proteins accession number begin with '77'
    • the original proteins accession number begin with '00'
  • the first two digits of the accession number are replaced for each (decoy) protein.

• Set Operations: the 'old' count elements has been reworked.
  • now 38 lists are supported, instead of ten in the previous version.
  • you can select which lists are used for counting by the use of checkboxes.
  • the selection of lists can now be cleared, i.e. all selected lists may be set to not selected.

• Database Search: added a new tool to the Tools menu: Database Search...
  • performs very fast searches in protein databases.
  • open fasta database: File menu | Open FASTA File...
  • an index for the protein database is automatically built
  • the index file can be saved for later searches: File menu | Save SuffixArray File...
  • alternatively open index file directly: File menu | Open SuffixArray File...
  • in the text area Peptides Input amino acid sequences can be entered
  • after pressing the start button a search in the protein database index is performed and the result is displayed in the java console. In the Output text area only the number of successful hits is displayed.
  • in the Status text area internal messages are displayed, e.g. time needed for searches and building the index structure arrays.

• Modifications: removed the Phospho II modification, because it is a combination of phosphorylation and beta-elimination, which are both present in the modifications database.
• Modifications: changed the Full Name of S-Ethyl from S-Ethylcystine from Serine to S-Ethylcysteine from Serine.

• Modifications: added new custom modifications to the modifications table:

  • Code Name	Full Name	Mono mass	Avg. Mass	Sites	Composition
    Phospho II	Phosphorylation II	97.976896	97.9952	S T Y	H(3) O(4) P
    S-Ethyl	S-Ethylcystine from Serine	44.008456	44.1188	S T	H(4) C(2) O(-1) S

• Set Operations: the frame has now two pages (tabs):
  • 3 sets: all: the "old" content described earlier: 3 input areas, 4 different set operations and 4 output lists.
  • 10 sets: count elements: this tabbed page is new. There are 10 input areas working like the other 3 input lists. If the start counting button is pressed on the right hand side a table is built where the number of occurrences of all proteins are displayed. Each protein is only counted once per input list. The table can be sorted according to count or accession.

Change log (January, 6th, 2006):

• Set Operations: More set operations have been added which are selectable by radio buttons:
  • Intersection: display only proteins that are common between the input sets
  • Union: display all proteins that are in either of the input sets
  • Difference: display only those proteins that are appearing in the first list, but not in the first
  • Symmetric difference: display only unique proteins that are not appearing in any other list

Change log (December, 22th, 2005):

• Set Operations: For better comparability with the output of the set operation, the input files are also sorted lexicographically. Redundancies in the input files are filtered and the number of elements and removed redundancies are displayed in the text area header. Empty rows are not counted as redundant elements but they are also removed.

Change log (December, 20th, 2005):

• Set Operations: the 'old' Intersections tool has been renamed to Set Operations. To support additional operations beside intersection the union of two or three lists is also supported. Simply select the desired operation by the appropriate radio button. These are found just above the start and clear button.

Change log (December, 19th, 2005):

• Intersections: a new tool has been developed: Intersections. Basically it compares two or three lists and displays their intersections, i.e. the elements that are common between those lists.
  • if two lists are entered then only the intersection between those is calculated:
    • Input1 ∩ Input2
  • if there are three lists (input1, input2 and input3) then four intersections are calculated:
    • Input1 ∩ Input2
    • Input1 ∩ Input3
    • Input2 ∩ Input3
    • Input1 ∩ Input2 ∩ Input3
  • on the left side of the frame three input text areas are displayed. On the right there are four text areas for the resulting intersection lists. In the bottom left corner two buttons reside:Start and Clear. A push of the first obviously starts the calculation, while the latter button clears all text areas.
  • the lists may have any textual content, e.g. accession numbers or protein sequences, as long as each element is listed in a single row.
  • Three files may be opened via the File menu to fill the input fields.
  • Copy and paste of lists into the input fields via key strokes is also permitted.

Change log (November, 15th, 2005):

• FindPairs: the calculation of the error percentage was too complaisant. This has been corrected and the percentage value is now multiplied by 100.

Change log (September, 29th, 2005):

• Gathered a list of publications where Peakardt has been used. If there are any publications, posters, talks, etc. that are not on the list please let me know.
• New web design for the Peakardt site. Move your mouse to the left to let the new menu open. Select the desired menu item and the page will be displayed. If you scroll down to read further parts of the page the menu will nevertheless be there to open, if you move your mouse to the left border.

Change log (September, 25th, 2005):

• FindPairs: if pairs were found using sequence information, the appropriate peptide sequences are also underlined to be better noticeable, if a black and white colored screen shot is to be taken.

• FindPairs: the deleting of rows in pairs or proteins tables produced some minor calculation errors for the found pairs, outliers or the protein overlap. These have been eliminated.
• FindPairs: the statistical calculations of found pairs on spectra and protein level has been unified, fixing some minor accuracy issues. The overview for each spectrum carries the number of found pairs and outliers, if appropriate, for this particular spectrum. The overview for the whole protein carries all merged pairs found from all spectra. The number of outliers reflects all outliers from all spectra, these were not merged and just the sum of all outliers from the spectra.
• Properties: fixed a bug that prevented that the status of the checkbox filter outliers could be loaded from a properties file.

• FindPairs: if sequence information is used for quantitation, then the m/z value of the detected monoisotopic peaks of a peptide pair is displayed as Mass 1 and Mass 2. If this peptide pair is found in multiple spectra then only the masses from the first are displayed. Previously the masses of the theoretical peptides were shown.
• FindPairs: in the protein table the column for the charge state of the 'heavy' labeled peptide is omitted, if the charge state of both peptides of one pair should have the same charge state. The remaining column is now titled Charge State.
• FindPairs: if the different secondary modifications property is used this works now also for modifications at multiple sites of a peptide, e.g. if a peptide contains more than one lysine, missed cleavages are allowed and ICPT is used as labeling technique.

• Spectra Display: besides the mass of a peptide peak now its charge state is also displayed in the spectra view, if the peak labels should be visible.
• FindPairs: if the quantitative difference between two differently labeled samples is to be determined, two modifications have to be selected as secondary modifications and in the Properties | Modification tab the checkbox use different secondary modifications. If this checkbox is selected then for every selected secondary modification another theoretical peptide is generated, whereas if the checkbox is not selected then only one peptide is generated with all secondary modifications attached (if also multiple modifications is also checked).

• FindPairs: removing rows in the found pairs table for a spectrum now also forces a recalculation of the protein results table, if no sequence information was used.

• FindPairs: copying of the pairs and protein table can now be done by right clicking in the table and choosing the copy table menu item. The appearance of the table entries, i.e. locale conform formatting of the decimal numbers, arrangement and sorting of columns as well as the table header, are preserved in the clipboard and can be pasted into a spread sheet (e.g. Excel) or text processing application (e.g. Word) with no further formatting.

• FindPairs: in addition to the calculation of the absolute standard deviation (root mean square error), the relative deviation is calculated to better reflect the deviation to big expectation values and to make quantitation calculations more comparable. The relative deviation is the quotient of the expectation value and the standard deviation.

• FindPairs: the peaks of an isotopic pattern of a peptide have to share the same charge state as the monoisotopic peak. Otherwise they are not considered for quantitation.

• FindPairs: if a peptide pair is found multiple times in one spectrum, then in the column Spectra Indices in the Protein tab the count is now shown in parentheses behind the spectrum index. This greatly enhances the readability of the spectra list where this peptide pair is detected. Nevertheless you should investigate why this peptide pair occurs multiple times in this spectrum, because this might be due to too much background noise or too high mass tolerance. Unwanted peptide pairs may be deleted from the list to
• FindPairs, Properties: especially for LC/MS runs a new merging constrain for peptide pairs has been implemented. If the checkbox only merge pairs from adjacent spectra is selected, then peptide pairs with the same attributes, i.e. the same sequence or the same mass of the 'light' and 'heavy' peptide, which are found in different spectra, are merged in the protein result table only when the spectra are adjacent. Adjacent means that the spectrum index of the first spectrum is only one higher or lower than the second.

• FindPairs: corrected calculation of arithmetic mean and median. Both values now represent the ratio of the 'light' vs. the 'heavy' labeled peptide variant, i.e. Intensity1:Intensity2.

• FindPairs, Properties: For improving the statistical robustness the outlier filtering can now be done with either the arithmetic mean or the median as expectation value.

• Open [PMF, PMF recursively, PFF] : Peakardt now supports import of spectra files in Analyst 1.1 format. The original Analyst 1.0 files are still supported. The different file formats are automatically detected. Be aware of the fact that the columns may not be changed from its standard order prior to exporting the spectrum, because the changed values may not be imported correctly. Remember also to add the ending _qstar to the filename. Following are the column headers from the Analyst spectrum file assigned to the columns in Peakardt
  • m/z (amu) => m/z
  • Intensity (counts) or Peak Area => Intensity
  • Charges => Charge State

• FindPairs: to improve the quality of measurement extreme values which do not fit into the expected values can be automatically filtered. Intensity ratios are called outliers, if they are too far away a predefined range around the expectation value (i.e. arithmetic mean). The border is defined by a constant factor multiplied by the standard deviation. The larger the factor, the more intensity ratios are present within the range, the higher the deviation may be.

• ¹⁵NShiftCalculation: added support for selenocysteine into the calculation of ¹⁵N shifts, because protein databases, e.g. the NCBInr, contain proteins with selenocysteine. This further improves the prediction accuracy.

• Properties | FindPairs: a new checkbox has been introduced: merge pairs with different charge states. If this checkbox is selected, peptide pairs from different spectra are regarded as representing the same peptide even if they have different charge states. Otherwise these are handled as different peptides and the intensities are not summed up.
• FindPairs: if a pair (i.e. row) from the FindPairs output-table is deleted then the the Proteins tab is automatically recalculated, so that is no longer a difference between the found pairs is present.
• Modifications: added support for the new labeling reagent ICPT. These are the modification values:
  • ICPT light:
    • Code Name: ICPT light
    • Full Name: Isotope Coded Protein Tagging light label
    • Mono Mass: 140.094963
    • Avg. Mass: 140.1
    • Sites: K
    • Composition: C(7) O N(2) H(12)
  • ICPT heavy
  • Code Name: ICPT heavy
  • Full Name: Isotope Coded Protein Tagging heavy label
  • Mono Mass: 143.1050275
  • Avg. Mass: 143.1
  • Sites: K
  • Composition: C(4) C13(3) O N(2) H(12)

Change log (August, 9th, 2005):

• File | Open PMF: added a new filter for importing spectra which were exported from Excel. Details will follow.
• Find Pairs: if no spectra were found in a FindPairs search, then the arithmetic mean and the according error rate could not be calculated correctly. These spectra now no longer contribute to the overall protein quantitation calculation and are not displayed in the proteins table and do not affect the statistical calculations.

• FindPairs:a major update has been made to FindPairs. Finding pairs in different spectra was just the first step, but now whole proteins can be automatically quantified.
  • after the find pairs search the results from all spectra are gathered and displayed in the new Proteins tab in the FindPairs-frame. The tab is divided into three vertically arranged sections. Its content is depending on the presence of sequence information or if just mass shifts should be detected.
  • Just a mass shift (without sequence information):
    • on top there is the entered mass shift displayed.
    • in the middle the table with gathered information from the spectra is displayed in five columns:
      • Spectrum Index : the index number of the spectrum and spectrum table
      • Arithmetic Mean: for all found pairs in this spectrum
      • Root Mean Square Error: the belonging error rate
      • Find Pairs Time: the time in milliseconds needed for execution
      • Pairs Found: the number of all pairs found in this particular spectrum
    • the bottom section displays overview information of the whole search:
      • the overall arithmetic mean for all spectra
      • and the overall root mean square error
  • with sequence information (and the appropriate labeling modification)
    • on top there is the amino acid sequence of the protein displayed
      • if regulated peptides could be detected in the spectra, the appropriate subsequences of the protein are displayed in the color blue.
      • after the sequence following information is displayed:
        • sequence length: number of amino acids of this protein
        • amino acid overlap: the number of amino acids of all found peptide pairs
        • ratio: the amino acid overlap ratio in percent: amino acid overlap/sequence length
      • in the center the resulting table is displayed with information of the find pairs search containing eight columns:
        • Peptide Number: a continuous index of the found peptides
        • Peptide Sequence: the sequence of the 'heavy' labeled peptide (with associated modifications, if selected)
        • Mass 1 Single Charge: the theoretical single charged mass of the 'light' labeled peptide
        • Mass 2 Single Charge: the theoretical single charged mass of the 'high' labeled peptide
        • Intensity 1 Summation: if more than one peptide pair was found the absolute intensities of the 'light' labeled peptide are summed up and displayed in this column. Otherwise the intensity value of the sole peptide is displayed.
        • Intensity 2 Summation: if more than one peptide pair was found the absolute intensities of the 'heavy' labeled peptide are summed up and displayed in this column. Otherwise the intensity value of the sole peptide is displayed.
        • I Ratio I1:I2: the intensity ratio of the light vs. the heavy peptide intensity sums is calculated and displayed.
        • Spectra Indices: the list of spectra numbers where this particular peptide could be found. If one index appears more than once this means, that this peptide was found in different charge states in the spectrum.
      • on the bottom there is a statistical protein regulation overview of the find pairs search:
        • the overall arithmetic mean of the intensity sum ratio of all peptide pairs is displayed
        • the overall root mean square error of all peptides pairs
  • As in the FindPairs output-table in both new Proteins-tables a simple right click on a row opens a menu where the selected row can be deleted. The appropriate values like overall arithmetic mean, overall root mean square error and in case that sequence information is present the amino acid overlap and overlap ratio as well as the sequence coloring are recalculated.

• Decoy Database Builder: added the possibility of generating a combined decoy database.
  • if this option is chosen, a database is built with three variants for every original protein:
    • reverse
    • shuffle
    • random
  • if the button combine is pressed, for every protein found in the left textfield those three variants are written to the right textfield of the Decoy Database Builder frame.
  • if the menu item combine & append fasta is selected then the afterwards selected fasta protein database is analyzed and for every original protein a copy of this protein and the three above mentioned variants are written to the resulting decoy database.
• added support for the 21st amino acid Selenocysteine and changed accordingly the aminoacids.cfg file.
  • This has impact on all parts of Peakardt which handle amino acid sequences, i.e.:
    • Digest
    • Fragment
    • FindPairs
    • Decoy Database Builder
    • Amino Acid Exchange
    • EvAl
  • The needed values for the amino acid are taken from this site.
    • amino acid: Selenocysteine
    • one letter name: U
    • three letter name: SeC
    • monoisotopic mass: 150.95364
    • average mass: 150.03
    • the hydrophobicity value is not known to me, so I set it to: 0.0
    • number of N atoms: 1

• Decoy Database Builder: the naming scheme of the filenames for the decoy database has been changed for IPI databases. The filename for those databases is built as follows:
  • 'ipi.' + ['decoy_'] + decoy_type + database_name + '.fasta'
    • 'decoy_' is added to the filename, if the original version of the proteins is also added to the resulting database. If only the decoy version of the protein is added 'decoy_' is left out.
    • decoy_type is one of 'reverse', 'shuffle' and 'random'
    • database_name is the original filename except 'ipi.' and '.fasta'

• FindPairs: long awaited, but finally here :-)
  Peakardt.FindPairs now offers the ability to delete unwanted peptide pairs from the FindPairs output table. Just do a right click on the desired table row, then a popup menu appears, here you can select delete row. The selected row is deleted and the overview text below the output table is automatically recalculated: pairs found, arithmetic mean and root mean square error.
• FindPairs: if the property allow only the same charge state for peak pairs is selected, only one column for the charge state of the peptides is displayed in the output table, because only peptide pairs with the same charge state are displayed.
• Properties | FindPairs: a rearrangement of filtering properties has been made to clarify the use of the FindPairs filter:
  • the panel consider intensities is now titled filter results: with or without sequence information.
    • here the checkboxes consider intensities
    • and allow only the same charge state for peak pairs reside
  • the panel choose best pair is now titled filter results: only with sequence information
    • here the checkboxes choose only the best pair
    • and filter same peptides are placed
• Properties | FindPairs: a new checkbox has been implemented: filter same peptides. If it is selected (the default state) then peptides with the same mass in the primary and secondary labeled variant are filtered and not displayed in the FindPairs output table, because they do not contribute to the quantitation analysis of the spectrum/sample.

• FindPairs: problem: a digest of a protein may yield peptides which are not labeled with the applied labeling method. This is for example true for the D4 Lysine labeling, if the peptide does not contain any lysine. This resulted in theoretical peptide pairs where both peptides have the same modifications and mass. The find pairs algorithm then tries to find this peptide pairs in the spectrum and displays the found peptide pairs in the output table. This in fact is no algorithmic problem, but a problem of the labeling method. To compensate the occurrence of peptide pairs with no mass shift, a new property has been introduced: filter same peptides.
• FindPairs: when clicking on the textfield where to enter the shift per peak, the containing text is automatically selected, allowing immediate overwriting with the desired mass shift.
• FindPairs: the mass shift for detecting peak pairs is now depending on the charge state of the 'light' peptide, if the mass shift is an additive constant, i.e. the amino acid mod checkbox is not selected. The value entered in the text field shift per peak is the mass shift to be expected for singly charged ions. If the ions are doubly charged the mass shift is halved, if ions are triply charged the mass is divided by three, etc.

• FindPairs: the entries of the output table can be selected arbitrarily. Now you can select single or multiple cells, rows, columns or portions of rows or columns. Just drag and drop the mouse for continuous selections. Hold down the ctrl-key (on Windows machines) for multiple selections.
• FindPairs: added charge state information for both peaks to the output table to have a visual control if the property allow only the same charge state for peak pairs is working properly.

• FindPairs: replaced the original result table, which was in fact an HTML-formatted text file rendered and displayed in the Pairs tab, by a real table. This is the basis for more improvements in the results appearance and advances in the algorithm and usability.
  The major appearance stayed the same: different columns depending on the availability of sequence information; coloring of ratios depending on the up and down regulation factor.
  A major benefit right now is that the found pairs (displayed in rows) can be sorted according to each displayed value (e.g. Intensity Ratio).

• Decoy Database Builder: as you know more than the usual 20 amino acids appear in FASTA formatted protein databases. For building a reverse or shuffle decoy database the additional characters B, U, X and Z are not filtered and also appear in the resulting decoy database. The filtering is still necessary for building random decoy databases.
• Decoy Database Builder: To distinguish the different types of decoy databases produced by the Decoy Database Builder the naming scheme has been changed:
  • the file extension .fasta of the original file is removed
  • an underscore and the type of decoy database (reverse, shuffle or random) is added
  • finally _decoy.fasta is attached to the filename if the original data is also present in the resulting database.
  • if only the modified sequences are stored only .fasta is attached .
• Decoy Database Builder: Because the length of a line in the FASTA format is not mandatory the desired number of characters per line can be entered in the Properties | General textfield labeled number of sequence characters per line. For example:
  • NCBI: 80 characters per line
  • IPI: 60 characters per line

• FindPairs: if the search is performed with sequence information, the output table now contains for each found peak pair the sequence of the secondary peptide, because the first peptide may not be labeled at all (in this case the 'light' label is no modification of the 'normal' peptide configuration and the 'heavy' label would not be displayed as it is a secondary modification).
• FindPairs: improvements in algorithm design lead to a further increase in processing time. Unnecessary searches and calculations are omitted.
• Properties | FindPairs: according to the changes made to the FindPairs algorithm the checkbox additionally allow only the same charge state for TM pairs (a matching pair was only found when both measured peaks had the same charge state as the theoretical peptide) has been renamed to allow only the same charge state for peak pairs.

• FindPairs: the FindPairs algorithm now supports checking for same charge state, even if no sequence knowledge is present. If the checkbox allow only the same charge state for peak pairs from the Properties | FindPairs tab has been selected, peaks with different charge states may no longer match, resulting an a reduction of the rate of false positive hits.

• Decoy Database Builder: added new menu items to the File menu:
  • reverse & append Fasta...: builds a reverse database and appends this to the original fasta content. Each original protein is followed by its reversed protein counterpart. The resulting file is saved with the same name as the original file with the ending _decoy.fasta
  • shuffle & append Fasta...: same as reverse & append Fasta except that a shuffled decoy database is built
  • random & append Fasta...: same as reverse & append Fasta except that a random decoy database is built
  • Exit: Use this to close the frame. In addition to the button in the upper right corner of the frame.

• Added more information about fragmentation principles to the fragmentation table. First column shows how fragmentation is handled in Peakardt right now. On second and third column are additional and somewhat contradictory information from Ioannis Papayannopoulos and John Cottrell.

• When dealing with amino acid sequences from databases notice that more than the usual 20 amino acid single character codes are used. Right now Peakardt removes all characters which can not directly be assigned to one of the 20 amino acids.
  • Here you can find an overview of the characters occurring in FASTA protein databases.

• FindPairs: Output tables could not be saved if the filename of the spectrum contained a complete file path. This issue could be resolved by replacing several characters in the filenames by underscores.
  • Here is how to save all HTML output tables at once:
    • select the menu item File | Save all find pairs output table as...
    • in the opened file chooser select the directory where to save all the html files
    • press Save
  • The files are saved in the selected directory. The filenames are the same as the paths of the appropriate spectra with some replaced characters.

• New Feature: Decoy Database Builder
  • The false positive rate of protein searches can be estimated by searching decoy databases containing entries with 'right' (original) and 'false' (new) protein sequences. The ratio of 'right' and 'false' protein hits is a measure of the false positive rate.
  • Decoy databases can be created from the original fasta database files in different ways:
    • reverse: the protein sequence of the original database is simply reversed.
    • shuffle: all amino acids from the original sequence are put in a random sequence.
    • random: the weight of the original protein is calculated and a random protein of the same weight (considering the given mass tolerance) is calculated with the same probabilities of amino acid occurrence as the original protein.
  • The Decoy Database Builder can be called by selecting the menu item Tools | Decoy Database Builder...
  • The opened frame contains:
    • File menu:
      • open Fasta...: a fasta file can be opened and displayed in the original textfield
      • reverse Fasta...: a reversed version of the database is saved in the same directory as the original database. To distinguish the databases '_decoy.fasta' is appended to the original filename.
      • shuffle Fasta...: same as reverse Fasta..., but a shuffled database is built.
      • random Fasta...: same as reverse Fasta..., but a random database is built.
    • Original textfield:
      • on the left side of the frame
    • buttons beneath the Original textfield:
      • reverse: a reversed database is built from the Original textfield and displayed in the Decoy Database textfield.
      • shuffle: same as reverse Fasta..., but a shuffled database is built.
      • random: same as reverse Fasta..., but a random database is built.
      • clear: removes all entries from both textfield's.
    • Decoy Database textfield:
      • on the right side of the frame

• Import of _qstar.txt files: now the localized import of QSTAR data files works as expected.
  If the file was produced on a computer with a german locale setting then the _qstar.txt file contained commas instead of decimal points. Selecting GERMANY as a locale in the General Properties now handles this issue and correctly parses the input file. Beware that choosing the wrong locale setting for an input file might result in loss of data if e.g. the decimal divider is misinterpreted.

• 2D Spectra: fixed some bugs which lead to wrong coloring of 2D spectra:
  • when using ternary difference for comparing spectra multiple merged peaks from one spectrum now lead to a maximum intensity of |1|. From this follows that the 2D spectrum only shows three different colors: blue (peak from the first spectrum), red (peak from the second spectrum), green (peak from both spectra).
  • when changing the display properties, instead of the displaying compared 2D spectrum the 2D spectrum of the first peak list was displayed.
  • the wrong color scale was automatically chosen, if the minimal intensity value equaled 0, what was the case for all spectra with no negative intensities.

• Added a new menu: Tools with two menu items to choose from:
  • ¹⁵N shift calculation - displays the ¹⁵N Shift Calculation frame as described earlier
  • Amino Acid Exchange - this was formerly placed in the main toolbar
• ¹⁵N Shift Calculation Frame: Built a tool for calculating the ¹⁵N shift value from a given Fasta database:
  • The frame mainly consists of a table with values needed for calculating the ¹⁵N shift for a Fasta database:
    • Name1: the one letter code of the amino acid
    • # N atoms: the number of the nitrogen atoms of this amino acid
    • 14N Mass: the mono isotopic mass for the amino acid containing only 14N atoms
    • ¹⁵N Mass: the mono isotopic mass for the amino acid containing only ¹⁵N atoms
    • Rel. Mass Shift: the relative mass shift from 14N to ¹⁵N variant of this amino acid
    • Absolute Count: how many times was this amino acid counted in the theoretically digested database?
    • Count %: the relative portion of this amino acid to the occurrence of all amino acids in the database
    • Abs.Mass portion of 100 AAs: the relative occurrence of this amino acid multiplied by its monoisotopic mass, i.e. the relative mass portion of this amino acid in a set of amino acids with the same composition as the processed protein database. If we had 100 amino acids the value in each cell denotes the mass which this amino acid contributes to the whole mass of all amino acids.
    • Rel.Mass portion %: the absolute mass portion of this amino acid divided by the sum of all mass portions gives the relative mass portion of this amino acid. This cell denotes the relative mass portion of this amino acid to an arbitrary (i.e. unknown) set of amino acids.
    • Resulting Mass Shift: this is the relative mass portion multiplied by the relative mass shift for this amino acid
  • The database can be opened in the File menu of the frame with the menu item Process Fasta...
  • Immediately after the database is selected a theoretical digest is started with the previously selected properties.
  • Above the table resides a toolbar where the digestion enzyme can be selected
  • Amino acids are counted from the digestion results, i.e. the resulting peptides.
  • Fast settings for the theoretical digest would be:
    • Digest | output format: deselect intro desc., peptides, extro desc.
    • Digest | sort results: select appearance and ascending
    • Modification deselect cysteine treatment and other modifications and select all partial modifications
    • From the frame's toolbar choose No_digest as digestion enzyme. Only whole proteins are counted and not further digested.
  • After the theoretical digestion of the protein database and calculation of the ¹⁵N shift table, a frame is displayed with the result of the calculations:
    • the name of the database
    • the ¹⁵N mass shift factor for this database
    • the time for the theoretical digest
    • the time for calculation of the shift factor

• AminoAcid Frame: in the amino acid table a new column is listed which denotes the number of N atoms in this amino acid.
• Properties: in the General tab a third radio button for choosing the right difference calculation method between two spectra has been added
• CompSpec: for comparing two spectra now there exists a third method for calculating the resulting intensities:
  • absolute - use the absolute intensities for calculating the difference between two spectra
  • relative - first calculate the relative intensities and the use them for calculating the difference
  • ternary - intensities from the first spectrum are set to 1, intensities from the second are set to -1, if peaks with same masses (within the mass tolerance range) appear in both spectra then the resulting intensity is set to 0.

  The ternary setting results in an easy way of determining if two spectra share the same peaks:

  • peaks unique to the first spectrum are drawn in blue,
  • unique peaks from the second spectrum are drawn in red
  • and peaks appearing in both spectra are drawn in green.

  The coloring of the peaks is visible when switching to the bubble style (Properties | Display) or in the 2D Spectrum view.

• ColorScale: added a second color scale which displays the intensity scale if the lowest intensity of a spectrum is not negative. According to the last mentioned rules.
• Spectra: automatic adjustment of the bubble coloring. If the lowest intensity of a peak is greater than zero (what it is in most cases, but not if it is a compare spectrum) then the bubbles may now have 1024 different colors instead of just 512.
  • The color range is as following:
    • relative intensity from 1.0 to 0.75: shades of blue
    • relative intensity from 0.75 to 0.5: shades of green
    • relative intensity from 0.5 to 0.25: shades of yellow
    • relative intensity from 0.25 to 0.0: shades of red
  • If the lowest intensity is negative then the coloring scheme is as it was before:
    • relative intensity from 1.0 to 0.5: shades of blue
    • relative intensity from 0.5 to 0.0: shades of green
    • relative intensity from 0.0 to -0.5: shades of yellow
    • relative intensity from -0.5 to -1.0: shades of red

• FindMasses, Data, SumSpec, EvAl Frames: the menu item Open PMF... was correctly replaced by Open PMF recursively...
• FindMasses: made the algorithm faster and less memory demanding
• FindMasses: if your masses are not found in the opened spectra there may be different explanations for this behavior:
  1. choose higher mass tolerance
  2. select search masses present in the spectra
  3. normalize to masses present in the spectra
  4. only peaks with intensities > 0 are considered
• FindMasses: now only spectra for which masses have been found are displayed. This results in a more readable output table if the searched masses are only present in a minority of spectra.

• 2DSpec: if the color scale display properties are changed the 2D spectra image is automatically updated.
• Spectra and 2DSpec: the color mapping of the spectra in bubble style and the 2Dspectra was accidentally switched to binary, i.e. the most intense was painted in blue and all others were painted in green. Now the correct shading of the color scale was restored to the peaks on a scale of 512 different color values.
• ColorScale: the color scale was not updated, when the color scale values in the properties changed. This is now fixed.
• 2DSpec: fixed a performance issue. When displaying a 2D spectrum the whole CPU was working at full capacity. Now the workload is much lower because of less redrawing operations.

• Java Webstart: changed some parameters and links to make the automatic check for and installation of the newest Java Runtime Environment more easy for the end user.

• Spectra display: increased usability for the spectrum view (still experimental):
  • if a spectrum is shown and you want to view only a small portion of it and not the whole mass range, then it is no longer necessary to change the mass range in the Properties.
  • Just press the mouse button on the level of the minimal m/z value in the spectrum, hold it and drag the mouse to the desired maximal m/z value and release the button. After that only the desired mass range is displayed. (right now this only works once, so no additional zooming is correctly displayed)
  • To return to the full view of the spectrum (that is mass range form 0 to the mass of the peak with the highest m/z value) simply do a double click anywhere on the spectrum.
• Spectra display: the memory usage of each displayed spectrum was minimized, so that more spectra (>>5000) can be opened at once.

• 2D Spectrum: added cross hairs following the mouse pointer to the 2D spectrum.
• Peakardt general: changed the status message when recursively opening PMF directories to open PMF directories and PMF files.

• Properties | Modification: the title for the display type radio buttons was renamed from display type for digestion output to modifications display type [also for digestion, fragmentation and findpairs]
• Properties | Modification: added a new display type for digestion output: no display. If for readability purposes you do not want the modifications in your sequence to be displayed choose this option. The display types are as follows:
  • no display: the modification is not displayed in the output
  • composition: the molecular composition is displayed
  • full name: the complete name of the modification is displayed
  • code name: a short version of the modification name is displayed
• FindPairs: If sequence information is used for the FindPairs algorithm then the peptide sequence also contains its primary modifications. Formatted using the composition, the full name or the code name of the modifications (set in Properties | Modifications).

• Properties and all parts of Peakardt displaying data: significantly increased the speed of applying changes in the display properties. Now only the visible part of the data is redrawn after changing the display parameters. The display speed is no longer bound to the amount of spectra opened. If e.g. several hundred spectra files were opened in Peakardt changing the visible portion of the spectra via display range now only takes the fraction of a second.
• FindPairs: added a new feature for use with PMF data. If the sequence belonging to the opened spectra is known and no modification is only primary modifications are (corrected March 15th) entered then the FindPairs output table lists all peptides found in the spectra which can be derived from the sequence.
  This obviously results in much redundant information in the output table:
  • Mass 1 and Mass 2 are the same
  • Primary Mass and Secondary Mass are the same
  • ΔM_m and ΔM_t are both = 0
  • Intensity 1 and Intensity 2 are the same
  • I Ratio (I2:I1) and I Ratio (I1:I2)are both = 1

  To sum it up: all peptides contained in the spectrum are listed with their sequence, their theoretical and measured mass, as well as the ΔTM value which gives an indication of the quality of the mapping.

• FindPairs: added a description field in the output table for the type of isotope detection (find all, find at least and find exact)
• FindPairs: added another column to the output table.
  • [#¹³C;#¹⁵N]: the two parts of the table cells contain the number of ¹³C or ¹⁵N isotopes used for calculation of the intensities of the found pairs

• FindPairs: implemented the constraint that both elements of the pair (i.e. the labeled and the unlabeled peptide) should have the same amount of isotopic ¹³C peaks. If one peptide has more peaks than the other only the lesser number of peaks is used for calculating the overall intensity.
  The higher the mass of a peptide the higher is the probability of the occurrence of ¹³C and ¹⁵N peaks. Therefore a fixed number of peaks for the whole mass range (e.g. 400 to 2000 da) would lose some information and would not reflect a correct quantitative measurement.

• FindPairs: in previous versions of Peakardt a fixed number of isotopes to identify a pair was mandatory. Now it is possible to choose if the exact number of ¹³C peaks and ¹⁵N satellites is required or if at least the entered number of peaks have to be found in the spectrum. The third option is to find an arbitrary number of isotopic peaks, i.e. to find all the peaks that are present in the spectrum. To sum it up, the new options are:
  • find all peaks possible to find in the spectrum
  • find at least the entered number of ¹³C peaks and ¹⁵N satellites, and use all found peaks for intensity calculation
  • find exact the entered number of ¹³C and ¹⁵N peaks and use this for calculation. Do not care about the other peaks.

• Added a new menu item: Open PMF recursively... : if this menu item is selected all PMF files found in the selected directory and all its subdirectories are opened. This is very useful for the handling of the FlexAnalysis file and directory structure.

• Enzymes: added a new enzyme which actually is a combination of Trypsin_strict and Cyanogen_Bromide. This could be used to simulate the sequential digestion with both enzymes. The enzyme is named Tryp_CB cleaves at lysine, arginine and methionine, proline is an inhibitor and it cleaves the c terminus.

• FindPairs: extensively expanded the parameter listing in the output tab. All those parameters affect the FindPairs search and may be adjusted to feed the individual needs:
  • the algorithm uses sequence information, if an amino acid is entered, no matter what values are entered for manual or predefined shift values.
    • if a sequence was entered then the settings for the digestion algorithm will be displayed after the FindPairs parameters
  • if no sequence was entered then the next values are determined by the predefined combo box.
    • Possible predefined values are:
      • NCBI - ¹⁵N
      • Arabidopsis - ¹⁵N
      • E.Coli - ¹⁵N
    • if there is no predefined value entered then the manually entered value in the shift per peak textfield is used for finding the pairs.
    • check the box amino acid mod? if the labeling method affects amino acids rather than whole peptides.
  • use range for monoisotopic peaks (discard lower and higher masses)
  • monoisotopic mass tolerance (used to decide if two monoisotopic peaks can build a pair)
  • other isotopic mass tolerance (used to decide if ¹³C and ¹⁵N peaks belong to a monoisotopic peak)
  • number of considered ¹³C peaks
  • number of considered ¹⁵N peaks
  • consider intensities (set this if monoisotopic peaks should always have the highest intensity (may not be true for masses above app. 1900 da))
  • up regulation factor (if higher than this value the intensity ratio for this pair is painted in green)
  • down regulation factor (if lower than this value the intensity ratio for this pair is painted in red)
  • choose only the best pair (this works only with sequence information. sometimes there are more than one possible found pair within the mass tolerance, but only the best matching pair is displayed)
  • additionally allow only the same charge state (if choose only the best pair is selected then this parameter ensures that only measured peak pairs are matched with theoretical peak pairs, if both share the same charge state

• FindPairs / Properties: in some rare cases the theoretical digestion of a protein yields peptides with same masses (within the range of the mass tolerance) but different charge states. Now the find pairs algorithm allows only the same charge state for TM pairs. If the so labeled checkbox in the Properties | Find Pairs tab is selected only theoretical and measured peptide pairs are matched against each other. This results in a further increase in accuracy.

• the Mascot Generic Format (MGF files ) for MS/MS import is now supported.
  • choose from the file menu open PFF and the appropriate MGF file, and all spectra contained in this file will be opened.
  • due to fact that intensities may have decimals in this file format, it sometimes will be necessary to multiply intensities with a constant factor. Check the Properties | General tab for the right settings (e.g. factor 10).

• Enzymes: Added a new cleavage enzyme to the default enzyme list: Unspecific.
  • cleaves may appear after every amino acid (ACDEFGHIKLMNPQRSTVWY)
  • Proline is no inhibitor
  • cleavages appear from the c terminal site

• FindPairs: added a new menu item and its underlying functionality: save all found pairs output tables. Click on this item, choose a directory and all the output tables from a find pairs search are saved as html files in one batch. The default filenames are the same as the name of the corresponding spectrum followed by an underscore ,the index number and the ".html" extension.
• FindPairs: if more than one pair matches the theoretical masses of a peptide, then the ΔTM value is painted blue. This is a hint to investigate further. The other pairs can be listed by deselecting the checkbox labeled choose only the best pair from the Properties | FindPairs tab.

• FindPairs / Properties: added to the Properties | FindPairs tab the checkbox choose only the best pair (matching one peptide sequence). The name reveals the function.
• FindPairs: Introduced a new measure for the quality of the found pairs when comparing with the theoretical values: ΔTM, which is calculated by the following formula: ΔTM = |PM-M1|+|SM-M2|. Where
  • PM is the mass of primary modified theoretical peptide,
  • M1 is the mass of primary modified measured peptide,
  • SM is the mass of secondary modified theoretical peptide,
  • M2 is the mass of secondary modified measured peptide
  • having a list of possible found pairs belonging to the same theoretical digested peptide the one with the smallest ΔTM value is the best one matching the theoretical values.

• FindPairs: rearranged the output columns. There are now three major groups with their sub columns:
  • measured
    • Mass 1
    • Mass 2
    • Δ M_m
  • theoretical
    • Primary Mass
    • Secondary Mass
    • Δ M_t
    • Sequence
  • results
    • Intensity / Peak Area 1
    • Intensity / Peak Area 2
    • I Ratio (I2:I1)
    • I Ratio (I1:I2)
• FindPairs: directly choose the enzyme with which you want to theoretically digest the given sequence. The enzyme-combo box has been placed right to the sequence textfield. You do not have to take the long way via the enzyme or digest frame.
• FindPairs: added a new column to the FindPairs output table: Δ M_t displays the mass difference between the theoretical mass of the labeled and unlabeled version of the peptide. The "old" Δ M column is now called Δ M_m. It displays the measured mass difference of the found monoisotopic peak pair.
• FindPairs / Properties: fixed a bug in FindPairs occurring when applying a mass window for discarding low masses (In Properties | Digest | discard masses smaller or equal). Now all theoretical digested pairs are used for searching the spectra.
• Properties Frame: added some missing clarifications to the Properties Frame:
  • Modification tab: the cysteine treatment and oxidation of methionine are primary modifications, whereas the ¹⁵N labeling is a secondary modification
  • Digest tab: the sort results option is also working for the FindPairs output table
• FindPairs:implemented the sorting feature to the output table of FindPairs.
  • sort by size: the output list is sorted according to the mass of the peak representing the unlabeled peptide (with primary modifications)
  • sort by appearance: sorted according to the peptides position in the protein sequence
  • this both works in ascending or descending order
• FindPairs: fixed the annoyance of automatic deselection of the ¹⁵N labeling modification, when changed properties were applied.

• FindPairs: if the a sequence was entered for FindPairs, the output-table has now three more columns in addition to the ones before (Mass 1 and 2, Δ M, Intensity / Peak Area 1 and 2, Intensity Ratio and the inverse ratio)
  • Primary Mass: the theoretical mass of the primary modified peptide (unlabeled)
  • Secondary Mass: the theoretical mass of the secondary modified peptide (labeled)
  • Sequence: the sequence of the peptide belonging to those masses
• FindPairs: the new FindPairs version is ready to use. The main difference is that it uses the sequence information to find the quantitative differences within the spectra. There is no need to guess the shift for a known protein, because this can be directly derived via the Digestor. This is all done automatically.
  Here are the step by step instructions:
  • open your PMF-file
  • check the right settings for
    • FindPairs: e.g. mass tolerance for monoisotopic peaks (i.e. to match the digestor-masses with the measured peaks), number of ¹³C peaks / ¹⁵N satellites
    • Filter: use mass tolerance for detecting the ¹³C and ¹⁵N satellite peaks
    • Modification: MH-value e.g. [M+H], primary modifications e.g. acrylamide or oxidized methionine, and most important the secondary modifications i.e. the labeling method like ¹⁵N
  • enter the sequence of the protein / peptide that should be quantified in the sequence field in the Pairs-tab of the FindPairs-Frame.
  • hit the find Pairs-Button for starting the automatic quantitation process

• Properties: The time delay when applying changed properties is vastly reduced, if no property changes occur related to data display. I.e. if you do not change any display properties or change the look and feel of Peakardt then the frames are not redrawn.

• Digestor: implemented and optimized the feature of Primary and Secondary Modifications (see change log of December 17th).
• Properties: Note that all opened windows of Peakardt are updated if apply or OK is clicked in the Properties window. This may result in some delay if high amounts of data are displayed (e.g. many spectra or long digestion output lists).
• Fragmentor: added the select [M] value ([M+H], [M] or [M-H]) to the intro text of the fragmentation table for informational purposes.
• Digestor, Fragmentor: the discard mass filter did not work correctly for multiple charged ions. The ions were filtered as if they were only singly charged. Now every ion is discarded according to its correct mass.

• Digestor: fixed the wrong annotation of amino acid modifications. Those were always displayed at the first amino acid and are now placed at the right position.

• Properties and Digestor: fixed an internationalization issue which occurred when trying to use numbers formatted in german style.

• Digestor: There is now an additional mode for handling the modifications within the in silico digestion.
  • In the Properties | Modification tab the user can select to use primary and secondary modifications. These modifications, and if they are of primary or secondary type, can be selected as always in the Modifications frame.
  • Primary modifications are those, which occur in all peptides (the original "unmodified" version), secondary modifications are the additionally modified versions of the "unmodified" peptides. For example acrylamide adducts or oxidation of methionine can be seen as primary modifications, whereas ¹⁵N- or iTraq- labeling can be seen as secondary modifications.
  • So far the usage of primary or secondary modifications only works for complete modification and not if all partial modifications are to be displayed. Also the multiple modifications option should be selected in the Properties.
  • Because I am not present at the MPC until Jan. 3rd the above mentioned function is not put on the server, because it was not intensively tested and could bear some minor bugs.
• FindMasses and corresponding Properties: implemented more functionality and information to FindMasses:
• now it is possible to select one of the masses to be searched for (in the properties tab) as the normalization value. This is in contrast to the last version where the mass with the smallest intensity was automatically chosen for normalization.
• for easy recognition the header of the normalization mass column is painted blue in the results table.
• beside the normalized values the original intensity values are displayed as an additional part of the results table
• the parameter overview for FindMasses now contains the selected normalization mass, its index in the properties frame and the used mass tolerance.

• Modifications and AminoAcids Frame: debugged the table handling of this frames.
• AminoAcids: revised the internal amino acid handling.
  The visual appearance of the amino acid frame changed from plain text to a real table. Now sorting is supported for the amino acid characteristics: one letter, three letter and full name, the monoisotopic and average mass as well as the hydropathy value.
• General Feature / Properties: unified the formatting of floating point numbers. Now the general appearance of those numbers can be adjusted by the user in Properties | General.
  • The formatting options are:
    • locale:
      choose what language specific number formatting you prefer (this determines e.g. if you want a decimal point instead of a comma, or which grouping character is chosen). Available languages/countries are:
      • CANADA, CANADA_FRENCH, CHINA, ENGLISH, FRANCE, GERMANY, ITALY, JAPAN, KOREA, SIMPLIFIED_CHINESE, TAIWAN, TRADITIONAL_CHINESE, UK, US
    • fraction digits:
      determine your maximal number of digits after the decimal point. Choose from 0 to 10
    • use number grouping:
      check this if you want to group your numbers at all, i.e. insert a grouping character every 3 digits before the decimal point.
  • this is used in all output html-tables, spectra tables and views
    
• FindMasses: a new feature of Peakardt has been implemented. The FindMasses component allows the user to find arbitrary masses in all opened spectra (hundreds or thousands possible). Up to ten different masses can be selected. If some of those masses are detected in a mass spectrum the corresponding intensities are listed and normalized to the lowest intensity value. In one concise table all information of the spectra is gathered and differentially displayed.
  To use the FindMasses functionality first enter the desired masses in the properties (tab FindMasses). Predefined as the first four values are the monoisotopic values for the 114, 115, 116 and 117 iTraq reagents. If text fields stay empty or have a zero value, those values are not searched for.
  To search for those masses, then open the FindMasses frame by clicking on the appropriate button of the main tool bar. In the following frame open your spectra the usual way (File menu) and click on the Masses tab. There the actual masses can be selected by the checkboxes labeled with m 1 to m 10.
  After hitting the find Masses button one overview table is generated from all opened spectra. It includes the spectrum number, its parent mass, the intensity value of the selected masses (if found) and the value of the mass tolerance used for searching.

• Data, CompareSpectrum, Find Pairs: added error handling code when copy and pasting new spectra via the clipboard. If the clipboard contains data in the wrong format, i.e. no peak list, then no spectrum is added and an error message is displayed.
• Fragmentor: completely rewrote the internal fragmentation handling and the according algorithms. This resulted in an immense performance increase. E.g. the theoretical fragmentation of Serum Albumin resulted in 1211 fragments (only b- and y-ions) about 200 milliseconds vs. 1000 milliseconds in the previous version.
• Properties | Filter: clarified the usage of peak filters. If the checkbox immediately apply selected filters after pressing apply in the properties frame is selected and any of the filters is selected then the filters are applied to the opened data. This is a one way ticket. No data can be restored once fallen through the filter. Spectra have to be reopened.
• FindPairs: fixed an issue that arose when opening more than 20 (or whatever value was specified in the properties for the number of displayed tabs) spectra at once and tried to calculate their regulation factor. In this case the wrong spectra for finding the pairs was used so the factors were displayed at the wrong position. E.g. on position of the 104th spectrum the values of the 33th spectra were displayed (Nevertheless the calculations for the 33th spectrum were correct).
• Modifications: completely revised internal modification handling. Visual appearance of the modification selection frame has changed for better readability and easier selectability purposes (it is also faster :-) ). For short summation of details take a look here.
• Modifications: added some new user defined modifications: 12 different iTRAQ variants for MS/MS.

• Digestor: corrected the mass calculation for multiple charged peptides.
  • The following formula is used: Mⁿ = (M ¤ n · H) / n
    • M is the mass of the amino acid sequence plus the mass of the termina
    • ¤ is a place holder for + (positive) and - (negative) charged ions
    • n is the charge state
    • H is the mass of one Hydrogen
    • Mⁿ denotes the resulting multiple charged mass of the peptide
  • if neutral peptides (i.e. not [M+H] or [M-H]) are selected there can not be multiple charged versions of these ions, so in the resulting digestion table only the single charge peptide masses are displayed.
  • in the output table it is indicated right after the sequence of the peptide, if it has additional hydrogen, loss of hydrogen or is neutral: ^[+n],^[-n] or ^[1]. Where ⁿ is the charge state.
• FindPairs: for every found pair additionally to the ratio (intens2 / intens1) now the ratio (intens1 / intens2) is calculated. The first value is for ¹⁵N/¹⁴N, the second is for ¹⁴N/¹⁵N ratios.
• Digestor: the digestion output is complemented with state of the parameter consider intensities.
• Digestor/Fragmentor: The mass window as specified in the Properties works now for multiple charged ions, too.

• FindPairs: pairs can now be found without display of the output table. The table can be saved directly to an html file. This results in an immense speed up if the saved file is displayed in the internet browser of choice (seconds vs. minutes with several thousand pairs). Use the menu item Find Pairs and Save Table as ...
• Peakardt: switched from Java 1.4.2 to Java 5. Due to many internal optimizations the overall performance has increased and the new look of Peakardt is much 'cooler'. If Peakardt does not start properly you have to update to the new java version. Download JSE 5.0 JRE on the following page and install it with administration rights on your pc.
• Digestor, Properties: If a FASTA file is directly digested (menu item digest FASTA), the automatically added extension is ".digest" if the output format is dta style and "_digest.html" otherwise. The first may be viewed with an text editor the second is preferably viewed with your favorite internet browser.
• Properties | Digestion | output format: the display of the digested peptides can now be switched on or off. This is important if you want to digest whole databases or large protein lists and do not care about the composition of the digested peptides, but their number and their amino acids count.
• Digestor: all modification types should now work flawlessly. That means: complete modification and all partial modifications with or without multiple selected modifications.
  • complete modifications: for one peptide two versions are displayed: the completely unmodified peptide and another peptide with all selected and appropriate modifications.
  • all partial modifications: the unmodified peptide is displayed and also peptides with all possible combinations of modifications.
  • Beware: the number of possible combinations is exponential
    • let n be the length of the sequence and m be the number of possible modifications for each amino acid in the sequence
    • the number of digested modified peptides is 2^n^m = 2^n·m
    • e.g. for a sequence of ten amino acids with 2 modifications each there are more than one million different modified peptides
    • e.g. a sequence of 64 amino acids with 4 modifications each will result in an number of peptides which is almost as large as the number of atoms in the universe.
    • a small calculated table with the number of possible peptides is here
• Properties | Modification: The default value for displaying the modification is set from composition to code name. The digestion output is now more readable.
• Digestor: the mass of all undiscarded peptides and the single protein weight of undiscarded peptides is only displayed in the digestion extro if at least one peptide is shown.
• Digestor: in the digestion intro text for the selected modifications not only the full name is displayed, but also the selected display type, i.e. the composition, full name or code name of the modification, as it will be displayed in the digestion output table.
• Data, FindPairs and CompareSpectra Frame: if more than one file is opened, the last opened file/spectrum/table is displayed as selected tab, and not the first.
• OpenPFF for QSTAR: removed input dialog about charge state, which was obsolete. Peaks with unknown charge states are automatically replaced with two peaks, one with charge state 2, the other with 3.
• Data, FindPairs, SumSpec: for spectra/peak lists containing peaks with more than one type charge state, e.g. generated by the ABI QSTAR instrument, an additional column in the table view has been added. The new column is named "Charge State" and will display the charge state of every peak in the spectrum.
• Data, FindPairs, SumSpec: reworked the table handling.
  • Tables can now be sorted ascending as well as descending by consecutive clicking on the header of the desired table column.
  • A third click on the header restores the original sequence of the table.
  • By pressing the CTRL (or STRG) key while clicking on further column headers the corresponding column values are used as subsidiary sorting criterions.
  • The order of sorting is annotated by small triangles in the column headers. The vertex of the triangle indicates the sorting order; its dimension indicates the sorting priority.
  • The sequence of the displayed columns can by changed by simply dragging the table header to the desired position.
• FindPairs: speed increase also appears for displaying the find pairs output.
• Digestor: increased the display speed of the digestion output about 30 fold by various nifty programming tweaks :-)
• AminoAcids: completely reworked internal amino acid handling for processing acceleration purposes. Visible outcome is the AminoAcidsFrame (from the Properties menu). If on the Properties tab for Digestion the sort results options are chosen, the content of the amino acid frame is sorted accordingly.
• Properties: if instrument settings for fragmentation are selected, the selected instrument is now saved. The settings for the selected instrument overwrites the other selected ion series.
• Properties: corrected a small item length error while saving and opening properties configuration files.
• Digest and Fragment: added to the information about the digested peptide/fragment is the charge state of the ion. Shown in green superscript numbers as the last item on the line.
• Properties: the textfield for entering the charge state of the digested peptide is replaced by ten checkboxes. With this change it is now possible to select more than one charge state for digestion or fragmentation at once. Useful for e.g. ABI QSTAR and other instruments which produce different multiple charged ions.
• Digest: more information about the digestion parameters is displayed in the output:
  • all selected modifications
  • is complete modification or all partial modifications selected
  • multiple modifications or only one modification allowed per site
  • modify all peptides termina or only the protein termina
• Digest: Choose which information about the modification is displayed in the digestion output. Select from Composition (the previous standard), Full Name or Code Name of the modification.
• Digest: Completely reworked the digestion with Complete Modification switched on, because of some terminal modification mismatches. The digestion now works faster and more accurate.
• Digest: disabled induce fragmentation checkbox, because fragmentation following the digestion is not implemented right now
• Fragment: added more information about parameters to the fragmentation results intro: i.e. which charge state, is used for calculations; which ion series are selected for fragmentation and if the first or the last of the n- or c-terminal ions is discarded
• Fragment: Disabled ion series which are not working correctly: internal, m ap, bp,bs, yp and ys ions.
• Digest and Fragment: Now there is a complete mass window for the digested and fragmented products. An upper and lower threshold can be entered in the Properties to discard greater or lower masses from the digestion or fragmentation output. Helps concentrating only on the products detectable by the method of choice.
• Every occurrence of the menu item "Save X" has been replaced by "Save X as...". It is always necessary to specify a directory and filename. Accordingly the menu item "Open X" has been replaced by "Open X..." following the GUI design rules for software development."X" may be any file format readable/writeable by Peakardt. This affects following frames: Main, Properties, Contaminants, Enzymes, Amino Acids, Modifications, Data, SumSpec, 2D Spec, CompSpec, FindPairs, Digest, and Fragment
• aaExchange: Removed the menu bar in the Amino Acid Exchange frame, because there is no need for opening and saving files here.

• Modifications handling has been reworked and updated: There are now two different files: the first (modifications.cfg) with the original modification database from Unimod (easily updateable with the obvious menu item: Web Update); the other (modifications_custom.cfg) with the custom modification for the lab. This file is a plain text file which can be altered with your favorite text editor.
• Custom modifications so far:
  • ¹⁵N labeling for one ¹⁴N atom
  • ¹⁵N labeling for two ¹⁴ atoms
  • ¹⁵N labeling for three ¹⁴ atoms
  • ¹⁵N labeling for four ¹⁴ atoms
  • Methionine oxidation
  • Cysteine iodoacetamide derivative
  • Applied Biosystems iTRAQ^TM at Lysine
  • Applied Biosystems iTRAQ^TM at Tyrosine
  • Applied Biosystems iTRAQ^TM at Nterm
• See the update log for details in the modification database changes. Following are just the full names of the modifications.
  • updated modifications in the database
    • FP-BiotinH
  • New modifications in the database:
    • 2 Hydroxyphenylglyoxal arginine
    • 6-aminoquinolyl-N-hydroxysuccinimidyl carbamate
    • Acrolein addition +112
    • Acrolein addition +38
    • Acrolein addition +56
    • Acrolein addition +76
    • Acrolein addition +94
    • ADP Ribose addition to Arginine
    • Applied Biosystems iTRAQ(TM) multiplexed quantitation chemistry
    • APTA d3
    • APTA- d0 with no neutral loss
    • APTA-d0
    • APTA-d0 with quaternary amine loss
    • ¹³C label (Arginine)
    • ¹³C label (Lysine)
    • ¹³C label (Phosphotyrosine)
    • ¹³C label (Tyrosine)
    • dimethyl labeling (D)
    • DiMethyl-CH2D
    • EAPTA d0
    • EAPTA d5
    • EDT
    • glycosylated asparagine
    • glycosylated asparagine 18O labeling
    • Hydroxyphenylglyoxal arginine
    • labeling Carbonyl Arg
    • Labeling Carbonyl Lys
    • Labeling Carbonyl Proline
    • Labeling Carbonyl Threonine
    • Michael addition of BHT quinone methide to Cysteine and Lysine
    • N-ethyl iodoacetamide-d0
    • N-ethyl iodoacetamide-d5
    • O18 label at both C-terminal oxygens
    • phosphorylation to amine thiol
    • Serine reduced to aminobutynate
    • Serine to Alanine
    • Shimadzu 12CNBS
    • Shimadzu 13CNBS
    • Threonine reduced to aminobutynate
    • Threonine to Alanine

• Digestor and Properties: multiple charged ions are now supported in Digestor. Enter the desired charge state of the peptides in the Properties frame and see the correct charged peptide masses in the digestion output list.

• Import of QSTAR-spectra and FindPairs: the import function has been optimized. The charge state does not have to be entered for every spectrum, if a peak with undefined charge state is discovered. Now Peakardt assumes for that peak that it may have a charge state of 2 or 3. To compute with this dual behavior for every "undefined" peak will be replaced by two separate peaks. The first with charge state 2, the second with 3. The other attributes remain the same. This causes the find pair algorithm to be slightly slower, because of the increased number of peaks.
• FindPairs: a mass range window is supported. If this option is selected only pairs within the lower and higher bounds are reported. This is useful for e.g. the MALDI matrix mass range or for iTRAQ analyses (mass range: 114 to 117)
• FindPairs: the algorithm is now able to detect an arbitrary number of ¹⁵N-satellites, not just only one. The compared intensities/peak areas are calculated from all selected peaks.
• FindPairs: after the results table there are now two more tables.
  • The first gives a short overview about the FindPairs search:
    • the number of found pairs
    • the arithmetic mean
    • the root mean square error
    • the time needed for the search in milliseconds
  • The second table lists the parameters used for FindPairs
    • the monoisotopic mass tolerance used to determine the matching pairs
    • the other isotopic mass tolerance used to detect the isotopic pattern for both monoisotopic peaks
    • the number of ¹³C peaks
    • the number of ¹⁵N satellite
• FindPairs: the display speed of the large result tables is now very much higher
• Properties and all other frames: the time needed for making all necessary display changes (after the OK or Apply buttons have been hit) is now significantly lower.
• Properties / FindPairs: a new parameter field has been introduced: mass tolerance for monoisotopic peak. This parameter works independently from the usual mass tolerance in the Filter tab. It is useful for finding the estimated modified/labeled peak in a broader range, without changing the accuracy for finding the C-isotopes or ¹⁵N-satellites. The usage can be switched on and off by selecting the checkbox.

Change log (since May, 14th, 2004):

• Properties / FindPairs: for more consistency the label changed from ¹⁵N to ¹⁴N ratio to number of ¹⁵N satellites. The following field is not in use right now.
• Properties / FindPairs: consider Intensities default value has changed from true to false, because the monoisotopic ¹⁵N-peak is not the most intense anymore, if the corresponding peak has more than 19 N-atoms (which can roughly be estimated for peptides with masses greater than 1500)
• FindPairs: The displayed results table can now be saved as an HTML-file.
• FindsPairs now supports searching in multiple spectra. Just open all the spectra you want to analyze. After pressing the find Pairs Button for each spectrum a tab with the resulting table is displayed. Use the parallel view from the Display properties to view more than one table at once.

• just a remark: FindPairs actually sums up the peak intensities / peak areas used for finding the isotopic pattern of 14N and ¹⁵N. If you for example select in the properties two ¹³C peaks and the ¹⁵N peak to be considered for pair detection, then every pair listed in the results table fits these requirements. The first intensity value (Intensity 1) is the summation of the monoisotopic 14N peak and its two carbon isotopes. The second value (Intensity 2) is the summation of the monoisotopic ¹⁵N peak, its two carbon isotopes and the ¹⁵N satellite. So what you select will be found, and what you found will be summed up.
• Digestor: the modification settings in the properties frame now work as a shortcut to the 'real' modification list. For convenience purposes the names of the buttons have stayed the same, but here is a table with the matching from button to the full name of the modification.

  AA	Button	Modification (Full name)
  cystein	acrylamide	Acrylamide adduct
  cystein	iodoactic acid	Iodoacetic acid derivative
  cystein	iodoacetamide	Cysteine Iodoacetamide derivative
  cystein	4-vinyl pyridene	S-pyridylethylation
  methionine	methionines oxidized	Methionine Oxidation
  all	15N Labeling	15N labeling for {one, two, three, four} 14N atom

  • if the buttons are pressed, the appropriate modification is selected from the list and vice versa.
  • selecting the ¹⁵N Labeling checkbox, will select all ¹⁵N modifications (for one, two, three and four atoms).
• Digestor: if more than one modification and complete modification was chosen, the mass of the modified peptide was not correctly calculated, because complete modification supposed only one modification. In the properties now multiple modifications (tab Modification) is selectable. If this is done all applicable amino acid modifications are considered. The radio buttons on the tab Modification should not be used right now. If modifications should be selected please do so by selecting them from the window opened by the menu item Modifications.... E.g. select Acrylamide adduct for cysteine, Methionine Oxidation for methionine and the ¹⁵N labeling for [1-4] 14N atoms modifications for the ¹⁵N labeling technique.
• FindPairs: time spent searching for the pairs is now shown in the output table
• FindPairs: output table heading now reads "Peak Area" instead of "Intensity" if the peak area is used instead of the intensity value. The right values were already used.

• FindPairs: reformatted the output table now each value has its own column, i.e. Mass 1, Mass 2, Δ Mass, Intensity 1, Intensity 2 and Intensity Ratio (Intensity 2: Intensity 1). Easy copying of the values to spread sheet software with no annoying ":" anymore.
• FindPairs: now already opened spectra are immediately displayed when FindPairs is opened.
• default charge state is now "1" instead of "0". This made problems with files which have no charge state defined (like Xmass and FlexAnalysis files)
• reworked opening of files. Choice of peak area instead intensity value works now for Xmass, FlexAnalysis and ABI QSTAR files.
• debugged file opening for Xmass and FlexAnalysis files.

• Properties: choose whether to use peak intensities or peak area as intensity value for further processing (i.e. Find Pairs).
  Tab: General, checkbox use Peak Area instead of Peak Intensity
  Note: only useful, if opened peak list contains peak area information (right now only ABI QSTAR)
• Properties: because some peak lists contain floating point intensities (even < 1.0), it is possible to automatically adjust the intensity values by multiplying by an arbitrary factor. Tab: General, textfield : multiply intensities with:
• FindPairs: consider charge state, when calculating the isotopic peaks.
  For ABI QSTAR every peak has its own charge state (or Undefined) every other supported peak list has only one charge state defined for the whole spectrum.

• added another column in the FindPairs output table for the mass difference of the found peaks (ΔM), to see the difference to a fixed peptide shift.
• now isotopic patterns in spectra can easily be found:
  • ¹³C isotopes with an arbitrary number of peaks can be found (most times only one or two distinguishable peaks are present, depending on the quality of the spectra)
  • additionally the ¹⁵N satellite can be found, adjustable to the pureness of the ¹⁵N (default is 95% ¹⁵N and thus 5% ¹⁴N)
  for better identification of single patterns the intensities of the peaks can be evaluated. If this applies, the monoisotopic peak should always be the highest. Remember that this setting will not show pairs with overlapping patterns.
• values for setting the right color for up up- and down regulation of found peptide pairs can be entered
• the arithmetic mean and the root mean square error (i.e. standard deviation) is calculated and displayed for the found peptide pairs
  • arithmetic mean is also shown in green or red, if appropriate
• added nice formatting to the resulting table of a FindPairs run
  • up regulated peptides are shown green
  • down regulated are shown red

• implemented novel FindPairs algorithm for automatically finding pairs of modified and unmodified peaks. Especially useful for all sorts of labeling (¹⁵N, ¹⁸O,...)
  • open your spectrum, enter the expected mass shift, choose if it is an additive or multiplicative shift (i.e. peptide or amino acid modification) and press the find Pairs button
  • the result is a table with the found masses and their corresponding intensity ratios
  • it is easily detectable, in which way the sequence represented by this peaks is regulated

• by choosing the appropriate instrument ion series are automatically selected
  • works for: MALDI-TOF, ESI-TOF, ESI-Q-TOF, ESI-ION-TRAP, ESI-3Q, ESI-1Q, ESI-FT-ICR, QSTAR, TOF-TOF, FOUR-SECTOR
  • clear all settings by selecting clear settings
  • note that d, v and w ion series are not implemented right now (for TOF-TOF or FOUR-SECTOR)
• copy and paste peak lists for fragmentor and digestor:
  • select dta style with no intro or extro description from the properties
  • Either digest proteins in the digestor or fragment peptides in the fragmentor
  • select copy from the Edit menu of digestor or fragmentor
  open the Data frame from the main toolbar
  • select paste new spectrum from the Edit menu of the Data frame
  • voilá, your digested / fragmented sequences have turned into a beautiful spectrum with all the possibilities of further processing
• implemented beta-version of fragmentor:
  • enter peptide sequence and get various fragmentation ion sequences:
    • a ions, a* ions, a° ions, a⁺⁺ ions,
    • b ions, b* ions, b° ions, b+° ions, b⁺⁺ ions,
    • y ions, y* ions, y° ions, y⁺⁺ ions, Y ions
    • c ions,
    • x ions,
    • z ions,
    • immonium ions
  • detailed information about fragmentation sequences
• digestor completed with
  • nice formatting of the induced modifications
  • reworked amino acids counting
  saving of output file in html format
• amino acid exchange: entering no sequence, but mass shift and protein sequence gives all protein subsequences with the entered mass.

Change log (since Nov. 28th, 2003)

• optimized loading of web files (less memory usage, less size restrictive, faster)
• import of a web database with almost 150 protein modification descriptions

• reworked import function for ABI QSTAR peak lists(read the updated instructions, still in german)
• saving of DTA-style digestions (just mass and how many peptides were found).
  • use this to view the mass distribution of a digest like a spectrum (with the possibility to manipulate (i.e. filter) it further)
  • this spectra view can handle more than the max. number of rows in a spreadsheet programs like Excel or Calc (there is a maximum of 65.536 rows)
• in digest extro-description the time for digestion process is displayed
• Digestor should now work flawlessly

• import ABI QSTAR peak lists (read the instructions in german)
• added hydropathic information to every amino acid
• created a No_Digest enzyme for calculating the mass of an intact protein (-list)
• added number of missed cleavages to digesting intro text
• saving of digested peptide list possible

• digestor now works with an arbitrary number of missed cleavages
• look & feel of different operating systems/window managers is selectable (i.e. Windows, CDE/Motif, Java Metal, MacOS)
• the mass tolerance algorithm has been optimized for handling relative intensities
• selection of parallel view of spectra, tables and data files or consecutive view (sub tabs)
• enhanced saving of spectra images
• wrote change log :-)