Protein Sequence Database Search

Which search method to choose?

In order to identify proteins from primary sequence databases, mass spectrometry data can be analyzed using following search methods:

• Peptide Mass Fingerprint

In this method, the experimental masses of peptides obtained from digestion of an unknown protein are compared to the predicted masses of peptides from the theoretical digestion of proteins in a database. If enough peptides from the experimental (usually MALDI MS data) and theoretical mass spectrum overlap, a protein can be identified. This method should be used when a sample to be analyzed contains a purified protein, and when a protein to be identified is from a species that is well represented in a sequence database. Although peptide mass fingerprint data continue to be accepted in the literature, the requirements have become more stringent. To confirm sequence of peptides identified from peptide mass fingerprint data or apply MS/MS ion search method for independent protein identification the APL also offers MS/MS fragmentation analysis on the most intensive peaks detected in MALDI MS spectra.

• MS/MS Ion Search

This method is much more powerful than the peptide mass fingerprint, because it uses peptide amino acid sequence information from MS/MS spectra. Experimental masses of peptide fragment ions are compared to the predicted masses of peptide fragments from the theoretical fragmentation of peptides in a database. If enough fragment ions from the experimental and theoretical mass spectrum overlap, a peptide can be identified. Obtaining matches to a number of peptides from a single protein provides a high level of confidence that the result is correct. This method is recommended for projects involving identification of proteins in complex mixtures or/and protein posttranslational modifications.

What protein sequence database should be used for searches?

The APL clients can request a search from any of the main protein sequence databases available publicly or a custom database when protein sequences are unique or constructed by a researcher. For large scale experiments, when results of additional statistical analyses have to be provided for publications (http://www.mcponline.org/misc/ParisReport_Final.shtml), additional searches may be done against a database in which the sequences have been reversed or randomized.