During a ProteinPilot software search, the Pro Group algorithm assembles all the peptides into their appropriate proteins. Before the protein ratios are computed, the peptides are automatically annotated according to specific properties that make them more or less useful for quantification. A key property that is directly determined during protein grouping is whether the peptides are shared between multiple proteins that have been detected in the sample. As these protein forms might have different expression changes, it is important to exclude these from the calculation of the final ratio for the protein.

A good example of this is shown in Figure 1 for an iTRAQ reagent – 4plex case, where multiple forms of the filamin protein were detected in the sample and both unique and shared peptides were identified. Filamin B has a much larger change in expression than Filamin C, as shown by the unique peptides. The computed ratio of the shared peptides is a blended ratio, as both protein forms contribute to its signal, and therefore should not be used to calculate the final ratio for the protein.

Other peptide features that are annotated are the confidence of the peptide (exclude peptides with <15% confidence); the signal strength of the peptide (exclude weak, low S/N peptides) and the feature probability of the peptide (exclude peptides such as semi-tryptics or low-probability modifications).

Only peptides with a check in the Used column in the Peptide Quantitation table are used in the calculation of the average ratio for the protein. The Annotation column text provides more information about why a peptide was or was not used for quantification.

The user can manually adjust these automatic selections by browsing for the protein on the Protein Quant tab and selecting or deselecting specific peptides using the Used checkbox. Average ratios, error factors and p-values are then recalculated for that protein.

Please refer to the ProteinPilot software help file for more detailed information on peptide quantification.