Ascore for phosphorylation sites
Researchers often have difficulty identifying the exact locations of phosphorylation sites on peptides with MS. So, Steven Gygi and co-workers at Harvard Medical School, Dartmouth–Hitchcock Medical Center, and Cell Signaling Technology, Inc., have developed an algorithm that automatically computes the likelihood that a phosphorylation-site determination is correct. The algorithm predicts several MS/MS scenarios on the basis of the possible phosphorylation sites on a given peptide. The theoretical data are compared with the experimental spectrum, and an ambiguity score (Ascore) is calculated. The Ascore method was validated on six data sets derived from phosphopeptides with known phosphorylation sites. The researchers also used the method to pinpoint phosphorylation sites on proteins in HeLa cells that were arrested during cell division. Many sites were determined with >99% certainty. In addition, the algorithm was more sensitive than Mascot or Sequest when used to localize phosphorylations. (Nat. Biotechnol. 2006, 24, 1285–1292)
Manual analysis emulator
Current search engines do not correctly identify all MS spectra, so researchers often resort to manual analysis to validate some of the results. But manual analysis takes a lot of effort and time, and consistent criteria do not exist. To automate this process, Katheryn Resing and colleagues at the University of Colorado (UC) at Denver, UC Health Sciences Center, UC at Boulder, and the Howard Hughes Medical Institute developed the manual analysis emulator (MAE). The tool uses the Sim score generated by the MassAnalyzer program to assess the likelihood of various fragmentation scenarios on the basis of known mechanisms. In addition, MAE calculates the proportion of ion current, a measure of the amount of the MS/MS ion current for which each peptide sequence is responsible. The program also has data-mining functions that allow researchers to determine whether a peak consists of data from multiple peptide ions. (Mol. Cell. Proteomics 2006, doi 10.1074/mcp.M600320-MCP200)
Mascot HTML and XML parser
To get the most out of their peptide identification results, researchers use a parser to extract data for additional analyses. However, many existing parsers are proprietary, expensive, or specific to certain platforms. So Raimond Winslow and co-workers at the Johns Hopkins University developed a parser that is open-source, free, and can be applied to data from any search engine.
The researchers demonstrated the parser with HTML and XML files that contain Mascot MS/MS and peptide mass fingerprinting search results. The tool creates a protein identification data object model, which is composed of objects called Protein, Peptide, Hit, Query, Modification, and Form. The parser extracts the relevant data for each object from the Mascot results and maps the relationships among the objects. This process reduces redundancy and provides information in a format that can be used by other programs to validate and analyze the data. Because of its modular design, the tool can easily adjust to accommodate new standards developed by the proteomics community. The source code is available at www.ccbm.jhu.edu. (Proteomics 2006, 6, 5688–5693)
Database tool for differential peptide expression
Mark Titulaer and colleagues at Erasmus Medical Center and FOM Institute for Atomic and Molecular Physics (both in The Netherlands) have developed a database application that allows researchers to easily store and compare mass spectral data from patients and controls. The tool includes a central database that contains metadata, such as sample codes. A file transfer protocol (known as FTP) server stores raw MS data, which can be exported to analysis software. The statistical program R also is included. The researchers demonstrated the application with MS data from breast cancer patients, prostate cancer patients, and their respective control groups. Although the tool was tested on MALDI TOFMS data, it also can store data generated by FTICR instruments. (BMC Bioinformatics 2006, 7, 403)
Return to Top
Copyright © American Chemical Society, 1155 Sixteenth Street N.W., Washington, DC 20036
