Abstract
Within this work I present a new protein identification strategy that overcomes the need for performing internal or close external calibration in MALDI-TOF-MS peptide mass fingerprinting. The strategy is based on the observation that the variation of peptide flight times, when measured on different positions on the sample support, are systematic and affect mainly the linear components (offset and slope) of the correlation between m/z and the square of the flight time. Consequently, the mass errors obtained when using a single set of calibration constants, determined at one position of the sample support, to calibrate all other time-of-flight spectra recorded from that support, are also systematic. The developed search algorithm recognizes these systemic trends in the mass errors, thereby allowing protein identification even with a low mass accuracy of the input data.
For the retrieval of the correct protein in a database search, I have developed a new scoring algorithm, which uses the parameters: standard deviation, number of matching peptide masses and the sequence coverage of the protein to calculate the score for each protein. Using this algorithm it was possible to correctly identify 52 out of 96 recombinant proteins of known identity, without any false identification. Moreover, I implemented the above identification strategy and scoring algorithms in a software package designated "MS-Proteomics", which automatically reads many peptide mass maps in a short time and performs all calculations for protein identification. For protein identification from 2D-gel electrophoresis, MS-Proteomics also comprises a 2D-gel viewer that links the search results to its corresponding spots on the gel image. All relevant results have been published or submitted for publication in the peer-reviewed scientific journals "Analytical Chemistry" and "Electrophoresis" (references are part of the appendix).
In addition, some of the results have been presented at the 48th ASMS Conference on Mass Spectrometry and Allied Topics, LA, California, USA, June 11-15, 2000. A web-based version of the program MSA 2.0 will be made available to the scientific community at http://www.scienion.de/msa, following publication of this thesis. |
