Peptide Sieve Package & Server
Important Note: PeptideSieve has been returned to public distribution now that it has been validated. Thank you for your patience.
It has been noted  that only a handful of a protein’s possible tryptic peptides are consistently observed in proteomics experiments. We denote these consistently observed peptides to be proteotypic peptides. Such peptides have a variety of potential applications in proteomic research including improving protein identification scoring functions of database search software, providing a panel of reagents for protein quantification as well as the annotation of genomes for coding sequences of e.g. the hundreds of sequenced bacterial genomes some of which are important model organisms in systems biology and a guide for peptide selection in targeted proteomics experiments. Here we present PeptideSieve, an alpha version of a computational tool to predict a peptide’s proteotypic propensity based on its physico-chemical properties. The resulting predictors have the ability to accurately identify proteotypic peptides from any protein sequence and offer starting points for generating a physical model describing the factors that govern elements of proteomic workflows such as digestion, chromatography, ionization and fragmentation.
The software consists of a single C++ program. The input is either a FASTA file of protein sequences or a TXT file of peptide sequences. The program then returns which of a protein's peptides are most likely to be proteotypic for each of four common experimental designs.
The program first performs an in silico digest of the protein and then converts each of the peptides into chemical property strings. The C++ program then computes a likelihood function, which scores the likelihood each peptide is proteotypic. It is important to note that the predictors are specific for particular experimental designs.
 Nat Biotechnol. 2007 Jan;25(1):125-31. Computational prediction of proteotypic peptides for quantitative proteomics. Mallick P, Schirle M, Chen SS, Flory MR, Lee H, Martin D, Ranish J, Raught B, Schmitt R, Werner T, Kuster B, Aebersold R.
Getting the software
New native C++ version (.51) released 5/2008: download the peptideSieve files from the Sashimi project at SourceForge. Linux, os x and windows binaries (PeptideSieve.exe PeptideSieve.linux.i386 PeptideSieve.osx.i386) are available.
A GUI windows version is available from our collaborator Chee-Hong! It is updated to PeptideSieve version .51
Build instructions are available here.
Running the software
PeptideSieve is a commandline utility. Running it sans arguments gives the usage instructions: Usage: peptideSieve [options] [files] PeptideSieve: Identify Proteotypic Peptides from a FASTA or TXT file. Version - 0.51 Options: -O [ --outputDirectory ] arg : set output directory -e [ --outputExtension ] arg : set extension for output files -o [ --outputFile ] arg : output file name if not input.extension -P [ --propertyFile ] arg (=properties.txt) : set property file -f [ --inputFormat ] arg (=FASTA) : FASTA or TXT, specifying input format -l [ --minSeqLength ] arg (=6) : minimum sequence length to consider -L [ --maxSeqLength ] arg (=40) : maximum sequence length to consider -m [ --minMass ] arg (=400) : minimum mass to consider -M [ --maxMass ] arg (=3000) : maximum mass to consider -c [ --numAllowedMisCleavages ] arg (=1) : maximum number of miscleavages to consider -s [ --saveConvertedFile ] : save the converted propertyFile -h [ --help ] : display usage information -d [ --experimentalDesign ] arg (=ALL) : which design to return, either NONE, ALL, PAGE_MALDI, ICAT_ESI, PAGE_ESI, or MUDPIT_ESI -p [ --pValue ] arg (=0.8) : only return peptides with p values greater than X
It is CRITICAL to either place the properties.txt file in the directory where PeptideSieve is being executed or to specify the location of properties.txt using the -P flag or PeptideSieve will work very strangely.