BMSorter: Yet another proteomics tool from Aebersold and Friends

Proteome analysis of Halobacterium sp. NRC-1 facilitated by the biomodules analysis tool BMSorter by Rueichi R. Gan1 at al. Molecular & Cellular Proteomics Papers in Press. Published on February 23, 2006 as Manuscript M500367-MCP200

BMSorter is a biological networks analysis tool that incorporates protein identification results into biological networks. BMSorter pulls together information from the identification of proteins with Trans Proteomics Pipeline (PeptideProphet and ProteinProphet) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) database (http://www.genome.ad.jp/). BMSorter is written in Perl language and interfaces via web browse (Apache web server). BMsorter can be obtained by contacting Wailap Victor Ng at wvng@ym.edu.tw

This paper reports on the systems analysis of the Halobacterium sp. NRC-1 soluble proteome identified by 2-dimensional liquid chromatography coupled with tandem mass spectrometry. Halobacterium is a class of archaea, found in water saturated or nearly saturated with salt. Proteins were identified using SEQUEST in combination with the Trans Proteomic Pipeline. BMSorter pulled the protein identification and the metabolic pathway information together. Cytoscape (www.cytoscape.org) was utilized display the protein identification information on amino acid metabolisms and the citrate cycle pathways in terms of the enzyme-metabolite interaction networks.

This title of this paper uses the somewhat novel term biomodules. Biological modules (or 'biomodules') are loose associations of preferred molecular interaction partners that interact to perform a collective function.
One simple and fun thing that was included is an equation to estimated number of false positive identifications from the Peptide Prophet results. The number of false positive identifications of N proteins with a probability value of P is equal to N x (1.0-P). Thus, if you identified 100 proteins all with a probability of .99 you can estimate that you have one false positive. In a realistic case the probabilities for each protein would be unique but the calculation remains trivial.

It is exciting to see how the average number of proteins that are identified in a standard MudPIT style experiment continues to grow. In this study 888 proteins were identified with a ProteinProphet probability (P) > 0.9. It seems that it was only a year ago that a study of this type might only identify a couple hundred proteins.