3. Instructions to run CLI for the identification of cross linked peptides

1- Update the configuration file (xLink.properties) according to your search settings.
Detailed information on the different parameters can be found on [here] (https://github.com/compomics/xilmass/wiki/3.1.-Xilmass-parameters-CLI).

2- Make sure that your database contains both target and decoy sequences with generic headers.
You can use [DBToolKit] (/projects/dbtoolkit) to generate your decoy database. Every decoy must contain either _REVERSED or _SHUFFLED next to the accession number position at its header. For example, one target header of PXXX has its decoy which is either PXXX_REVERSED or PXXX_SHUFFLED
For details, please see 5. Database.

3- In order to run Xilmass, execute the following line on the command prompt

java –jar xilmass.X.Y.Z.jar -c

Note that X.Y.Z stands for version number.

After running Xilmass, there will be a fastacp file on the given cxDBName. This search database is composed of all possible generated cross-linked peptides. The details on the presentation of this cross-linked peptide database can be found on here. There will be an output folder that is set by a user. In this folder, there are Xilmass outputs for ever mgf file, one file that contains validated cross-linked peptides-to-spectrum matches (XPSMs), one file that contains all XPSMs and one file for search settings.