Skip to content
COSS: CompOmics Spectral Searching
Java
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
lib/unknown/binary/AbsoluteLayout/SNAPSHOT
logs
resources
src
testData
.gitignore
README.md
README.txt
pom.xml
~$README.md

README.md

COSS


Project Description

COSS is a user-friendly spectral library search tool capable of processing large spectral libraries and supporting multiple file formats. COSS is developed in Java and hence it is platform independent. COSS uses external subsystems for file reading and writing. Next to file io subsystems there are three main processes: preprocessing, feature extraction and matching. All processes are organized and implemented in a modular fashion to facilitate future upgrades.

Go to top of page


Input Data

COSS supports the following spectral file format for query (experimental) spectra.

  • Mascot Generic Format: mgf
  • NIST spectral library format: msp
  • Institute of System Biology (ISB): mzXML
  • HUPO Proteomics Standards Initiative file format (PSI): mzData
  • ISB and PSI joint format: mzML
  • SEQUEST: dta
  • pkl
  • ms2

Accepted spectral library file formats are mgf and msp.

Go to top of page


Output Data

Users can export the result in excel. The output table contains 15 columns.

Parameter Description
Title Title of the spectrum (for msp file, Name is used as title)
Library Describes weather the library is true library or decoy
Scan num. Scan number
Sequence Peptide sequence from the matched library spectrum
Prec. Mass (M/Z) Precursor mass of query spectrum
Charge Charge of query specrum
Score Search score
Validation Validation, either 1% FDR or %5 FDR
#filteredQuerypeaks Query peaks after filtering the spectrum under 100Da mass window
#filteredLibraryPeaks Library peaks after filtering spectrum under 100Da mass window
SumIntQuery Sum of peak intensities in filtered query spectrum
SumIntLib Sum of peak intensities in filtered library spectrum
#MatchedPeaks Number of matched peaks found in query and library spectrua being matched
MatchedIntQuery Sum of peak intensities of query spectrum that have a match in library spectrum
MatchedIntLib Sum of peak intensities of library spectrum that have a match in query spectrum

It is also possible to save results in .cos format. This allows users to re-import previous results for visualization.

Go to top of page


Downloads

Download the latest version of COSS here.

COSS can be run with the user-friendly GUI or through the CLI.

Go to top of page


Usage

GUI

  • Download COSS from the provided link and unzip it.
  • On Windows you can run COSS by double clicking the COSS-X.Y.jar file. COSS can also be started from the command line using the following command:
$java -jar COSS-X.Y.jar

X.Y stands for the version number (eg. COSS-1.1.jar)
Make sure Java is installed on your machine.

  • Parameter Setting: Select and fill all required parameters.
  • Decoy generation: It is recommended to add decoy spectra to your spectral library for result validation. You can generate decoy library spectra using COSS builtin decoy generation. COSS has two algorithms to generate decoy spectra, reverse sequence and random sequence techniques. Click the GenerateDecoy menu and select the algorithm to generate the decoy spectra (which will be equal in size to your spectra library) and concatenate the decoys to your library.

*Note: make sure the spectral library is annotated before generating decoy library, if not you can use spectrum annotator provided in COSS.

  • Configuring File Reader: Click "Config Spec. Reader". At this time, the system disables the "Configure Reader" and "Start Seach" buttons until it is finished with the configuration.
  • Searching: Click "Start Searching", COSS starts searching and displays the status on the progress bar. The left-hand side window shows information of the query file. It also visualizes the spectra.
  • Result: To see the results, click on the Result tab from the main window. The upper table lists the experimental spectra while the lower table lists the top 10 matched spectra for the selected experimental spectrum. An interactive spectrum comparison view is presented at the bottom with the selected experimental spectrum (red) mirrored with the selected matched library spectrum (blue).

CLI

Command line searching is possible in COSS with the following commands:

java -jar COSS-X.Y.jar targetSpectraFile librarySpectraFile algorithm(0=MSROBIN, 1=Cosine similarity)   

or

java -jar COSS-X.Y.jar targetSpectraFile librarySpectraFile algorithm(0=MSROBIN, 1=Cosine similarity) precursorMassTolerance(PPM) fragmentTolerance(Da.) 

or

java -jar COSS-X.Y.jar targetSpectraFile librarySpectraFile algorithm(0=MSROBIN, 1=Cosine similarity) precursorMassTolerance(PPM) fragmentTolerance(Da.) maxNumberofCharge

Decoy spectra can be generated and appended with the following command: dV- reverse, dR- random

java -jar COSS-X.Y.jar -dV librarySpectraFile

Spectrum annotation can be done using the following command:

java -jar COSS-X.Y.jar -a librarySpectraFile fragment_tolerance

X.Y stands for the version number (eg. COSS-1.1.jar)
Make sure Java is installed on your machine.

Go to top of page


Java Maven Netbeans
java maven netbeans

Go to top of page

You can’t perform that action at this time.