PeptideAtlas software

From SPCTools

This is an evolving document about the PeptideAtlas code base, begun by Terry Farrah in June 2009.

The core data in PeptideAtlas is a large set of peptides organized into individual builds according to organism and sample type (for example, Yeast, Human Plasma, or Human Glycocapture). A peptide may belong to multiple builds.

$SBEAMS below refers to /net/dblocal/www/html/[devXX/]sbeams (as of June 16, 2009).
$PIPELINE refers to /net/db/projects/PeptideAtlas/pipeline.

1 Before a PeptideAtlas build
2 Creating ("building") a PeptideAtlas build
3 Loading a PeptideAtlas build
4 Displaying PeptideAtlas via web interface

[edit]

Before a PeptideAtlas build

In /regis/sbeams/bin is mostly stuff useful before building PeptideAtlas -- converting file formats and searching -- but some of it is useful during the build process: Mayu (used during step07 of build pipeline), calctppstat.pl (summary of TPP results on a single experiment), updateAllPaths.pl (used when moving pepXML and protXML files to new directories), showProteins.pl (summary of proteins in pepXML).

Mayu is the only one of the above called explicitly during the build process. Perhaps it should thus be stored in $SBEAMS or $PIPELINE, or at least linked there.

[edit]

Creating ("building") a PeptideAtlas build

A collection of csh, bash, and perl scripts, plus various text, XML, and csv files, all outlined in Building Peptide Atlas.

Many of these are in $PIPELINE:

$PIPELINE/recipes -- notes on how to create and load a build; not executable
$PIPELINE/run_scripts -- shell scripts for running the build pipeline
$PIPELINE/bin -- mostly perl scripts, called directly or indirectly from the scripts in $PIPELINE/run_scripts

Some of the software needed for the analysis step is in $SBEAMS/lib/scripts/PeptideAtlas: PAidentlist2mayu.pl, update_proteotypic_scores.pl, update_search_batch_statistics.pl.

[edit]