Common Types of False-Positives Identified in Shotgun Proteomics

摘要

False positives from protein sequence database search of MS/MS data remain a concern in proteomics research. Here we present five types of false positives identified by sequence alignment of MS/MS spectra. The false positives were caused during protein database search, because of misinterpretation of charge states, abnormal enzymatic digestion sites, misinterpretation of protein modifications, wrong assignment of modification sites, and incorrect use of isotopic peaks. While proved to be false positives, they have high statistic scores calculated from sequence alignment algorithm (e.g., Mascot), and can explain up to more than 80percent fragment ions present in the MS/MS spectra. Because of high sequence similarity between the false positives and their corresponding true hits, these false hits cannot be evaluated by the commonly used methods using reverse sequence database or scrambled database. A common feature of the false positives is existence of the unmatched peaks in the MS/MS spectra. Our studies therefore highlight the importance to use unmatched peaks to remove false positives and offer future direction to develop better sequence alignment algorithms for protein identification.