An evaluation of multidimensional fingerprinting in the context of clinical proteomics

Autor: Katherine Bonter, Eustache Paramithiotis, Heather Butler, Paul Kearney, Joanna M. Hunter, Patrice Hugo, Kossi Lekpor, Daniel Chelsky, Gregory J. Opiteck, Marie-Josée Benoit, Daniela Vasilescu, Michael Schirm
Rok vydání: 2007
Předmět:
Zdroj: PROTEOMICS – Clinical Applications. 1:457-466
ISSN: 1862-8354
1862-8346
DOI: 10.1002/prca.200600890
Popis: Multidimensional fingerprinting (MDF) utilizes measurable peptide characteristics to identify proteins. In this study, 3-D fingerprinting, namely, parent protein molecular weight, peptide mass, and peptide retention time on RPLC, is used to identify 331 differentially expressed proteins between normal and human colon cancer plasma membrane samples. A false discovery rate (FDR) procedure is introduced to evaluate the performance of MDF on the colon cancer dataset. This evaluation establishes a false protein identification rate below 15% for this dataset. Western blot analysis is performed to validate the differential expression of the MDF-identified protein VDAC1 on the original tissue samples. The limits of MDF are further assessed by a simulation study where key parameters such as database size, query size, and mass accuracy are varied. The results of this simulation study demonstrate that fingerprinting with three dimensions yields low FDR values even for large queries on the complete human proteome without the need for prior peptide sequencing by tandem mass spectrometry. Specifically, when mass accuracy is 10 ppm or lower, full human proteome searches can achieve FDR values of 10% or less.
Databáze: OpenAIRE
Externí odkaz: