The keratin contamination is almost negligible in the plasma sample treated with ProteoMiner, while for the sample depleted with ProteoPrep20, keratin peptides account for almost 10% of the total number. This can be ascribed to the very laborious procedure of plasma depletion that requires a heavy and time-consuming handling of the sample. Obviously,Syrosingopine to allow such a comparison, the operator, the quality of reagents, and the technical precautions were identical for both approaches and the 2 procedures were conducted in parallel. Moreover, the protocol suggested by the supplier of Proteo- Prep20 indicates, as an optional step, the precipitation of proteins prior to trypsin digestion and MS analysis. We clearly verified that such a procedure cannot be considered as optional because of the high amount of polymeric compounds released by the depletion column that strongly interferes with the MS analysis by suppressing peptide ionization. We could demonstrate that plastic contaminants are released into the sample not only from the depletion column itself,INI-43 but also when filtering the plasma at the preliminary step and when concentrating the final depleted sample with the provided concentrator. An example of base peak chromatograms obtained after each step of the depletion protocol applied to an ultra-pure water sample and a characteristic MS spectrum of the contaminants are shown in supplementary Figure S1. Table 2 reports the number of peptides, proteins and protein groups that were identified with at least one peptide and 99% confidence with Mascot, Sequest and the combination of the two. Our results show that Sequest clearly outperforms Mascot in terms of number of peptides identified. A manual screening of the identified peptides suggests that Sequest is able to identify more modified peptides and more often identifies the same peptides with different charge states, while Mascot generally fails to do so. However, by looking at the MS/MS spectra, it is possible to deduce that Sequest is less stringent in terms of spectral quality. A partial overlapping of the results could be observed, but an indepth statistical analysis should be performed in order to characterize the common features of the peptides that are better identified by one or the other of the search engines.