Techniques for Mass Spectrometry - Scalable Library Searching

摘要

In proteomics, traditional database searching techniques rely on comparing experimental mass spectra against an in-silico, generated theoretical mass spectra in a translated database from DNA sequences. However, since not all databases are complete, experimental spectra of mutant or novel sequences that are not in a database are missed in the identification schemes. This is, in part, is responsible to the low identification (at best 50%) of MS/MS spectra that are assigned a sequence for identification. Complementing database searching is spectral library searching. No theoretical MS/MS generation and searching is necessary since experimental data is the database. In general, experimental libraries are built from experimental mass spectra. Various library search algorithms are employed to identify new experimental mass spectra against what is already in the library. As these libraries grow the necessity for high performance searching algorithms are increasingly needed. This limits the utility of these libraries to sub-searches minimizing or reducing its global applicability.