2CEA, Universite Joseph Fourier, 17 Avenue des Martyrs, Bat. C3, 38054 Grenoble Cedex 9, France; fax: +33 (43) 878-5051
3Uppsala University, Institute for Cell and Molecular Biology, Box 596, BMC, SE 75 124, Uppsala, Sweden; fax: +46 (18) 471-7209
4Semenov Institute of Chemical Physics, Russian Academy of Sciences, ul. Kosygina 4, 119991 Moscow, Russia; fax: (495) 137-8247
5Emmanuel Institute of Biochemical Physics, Russian Academy of Sciences, ul. Kosygina 4, 119334 Moscow, Russia; fax: (495) 137-4101
* To whom correspondence should be addressed.
Received February 5, 2009; Revision received April 20, 2009
Generation of a complex proteome database requires use of powerful analytical methods capable of following rapid changes in the proteome due to changing physiological and pathological states of the organism under study. One of the promising technologies with this regard is the use of so-called Accurate Mass and Time (AMT) tag peptide databases. Generation of an AMT database for a complex proteome requires combined efforts by many research groups and laboratories, but the chromatography data resulting from these efforts are tied to the particular experimental conditions and, in general, are not transferable from one platform to another. In this work, we consider an approach to solve this problem that is based on the generation of a universal scale for the chromatography data using a multiple-point normalization method. The method follows from the concept of linear correlation between chromatography data obtained over a wide range of separation parameters. The method is further tested for tryptic peptide mixtures with experimental data collected from mutual studies by different independent research groups using different separation protocols and mass spectrometry data processing tools.
KEY WORDS: high performance liquid chromatography, proteomics, mass spectrometry