MS2PIP is a tool for predicting MS/MS signal peak intensities from peptide sequences. It can train prediction models from your data and/or predict fragment ion intensities for peptides.

MS2PIP employs the Random Foresttm regression algorithm (RF) as created by Leo Breiman and Adele Cutler. MS2PIP uses the CRAN R implementation.

MS2PIP exploits the potential of multi-core architectures through a forkmanager.

MS2PIP was developed and tested on Ubuntu 10.04.3 LTS.


MS2PIP is written in Perl. The following perl CPAN packages are required:

–          LWP::Simple
–          Parallel::ForkManager
–          File::Temp

These can be downloaded from MS2PIP requires an installation of CRAN R ( and makes system calls to the randomForest package
which can be found at (we used version 4.6-7).


code (contains example datasets)
Random Forest models used in the paper