Identification and characterization of proteins via database searching of tandem mass spectrometry data of peptides has become a central technique in proteomics. The massive amount of data generated precludes manual interpretation and validation of the identifications. The current standard procedure is repeating experiments and estimating false discovery rates by searching reverse or garbled protein sequence databases. One danger of this approach is that erroneous assignments can be made quite consistently and reproducibly by the software based on the measurement of a parent mass and the corresponding fragment ions. By derivatizing peptides the parent and fragment masses have to change consistently leading to non-redundant measurements. Here we explore chemical perturbation as a means of improving the confidence level of identifications.