Modern tandem mass spectrometry (MS) instruments combine features of fast duty cycle, exquisite sensitivity, and unprecedented mass accuracy. Tandem mass spectrometry, which is an ideal match for the large-scale protein identification and quantification in complex biological systems. In a shotgun proteomics approach, proteins in a complex mixture are digested by proteolytic enzymes such as trypsin. Subsequently, one or more chromatographic separations are applied to resolve resulting peptides, which are then ionized and analyzed in a mass spectrometer. To acquire tandem mass spectra, a particular peptide precursor is isolated, and fragmented in a mass spectrometer; the mass spectra corresponding to the fragments of peptide precursor is recorded. Tandem mass spectra contains specific information regarding the sequence of the peptide precursor, which can aid the identification of the peptide/protein.
Sequence database searching is widely used currently for mass spectra based protein identification. In this approach, a protein sequence database is used to calculate all putative peptide candidates in the given setting (proteolytic enzymes, miscleavages, post-translational modifications). The sequence search engines use various heuristics to predict the fragmentation pattern of each peptide candidate. Such derivative patterns are used as templates to find a sufficiently close match within experimental mass spectra, which serves as the basis for peptide/protein identification. Many tools have been developed for this practice, which have enabled many past discoveries, e.g. SEQUEST,3 Mascot.4
Due to the complex nature of peptide fragmentation in a mass spectrometer, derivative fragmentation patterns fall short of reproducing experimental mass spectra, especially relative intensities among distinct fragments. Thus, sequence database searching faces a bottleneck of limited specificity. Sequence database searching also demands vast search space, which still could not cover all possibilities of peptide dynamics, exhibiting limited efficiency post-translational modifications). The search process is sometimes slow and requires costly high-performance computers. In addition, the nature of sequence database searching disconnects the research discoveries among different groups or at different times.
First, a greatly reduced search space will decrease the searching time. Second, by taking full advantage of all spectral features including relative fragment intensities, neutral losses from fragments and various additional specific fragments, the process of spectra searching will be more specific, and it will generally provide better discrimination between true and false matches.
Spectral library searching is not applicable in a situation where the discovery of novel peptides or proteins is the goal. However, more and more high-quality mass spectra are being acquired by the collective contribution of the scientific community, which will continuously expand the coverage of peptide spectral libraries.
For a peptide spectral library, to reach a maximal coverage is a long-term goal, even with the support of scientific community and ever-growing proteomic technologies. However, the optimization for a particular module of the peptide spectra library is a more manageable goal, e.g. the proteins in a particular organelle or relevant to a particular biological phenotype. For example, a researcher studying the mitochondrial proteome will likely focus on analyses within protein modules within the mitochondria. The research community focused peptide spectral library supports targeted research in a comprehensive fashion for a particular research community.
Domokos, L., Hennberg, D., and Weimann, B. 1984. Computer-aided identification of compounds by comparison of mass spectra. Anal. Chim. Acta 165:61-74. ↩
Yates, J.R., 3rd, Morgan, S.F., Gatlin, C.L., Griffin, P.R., and Eng, J.K. 1998. Method to compare collision-induced dissociation spectra of peptides: Potential for library searching and subtractvie analysis. Anal. Chem., 70:3557-3565. ↩
Eng, J.K. et al. (1994) An approach to correlate tandem mass-spectral data of peptides with amino-acid-sequences in a protein database. J. Am. Soc. Mass Spectrom., 5,976-989. ↩
Perkins, D.N. et al. (1999) Probability-based protein identification by searching sequence database using mass spectrometry data. Electrophoresis, 20, 3551-3567. ↩