Table 2.

Proteases predicted by in silico protease mapping to peptide markers included in the CCA-specific bile and urine proteomic models. Analysis was carried out for bile and urine peptide markers independent from each other. Differences in the predicted activities between the case and control groups are represented by the fold change between the ion signals of the protease associated peptide substrates in the case and control groups. Fold change calculations are based on the following principles: Ion signals for the CCA peptide markers are extracted from the normalized peptide lists of individual samples as described in the methods section. Protease prediction is performed on the set of CCA peptide markers for each individual patient. CCA peptide markers for which the N- or C-terminal amino acid sequence motif could be attributed to the same protease were integrated by their ion signals to calculate the proteolytic activity of that protease in one patient. In a final step, proteolytic activity of a protease is integrated over all patients of one group and group wise compared to generate the final list of differences in protease activities based on the peptides ion signal distributions in the case versus the control patient group. P values were calculated by the Mann Whitney U test. Abbreviations: BBD benign biliary disease, CCA cholangiocarcinoma, PSC primary sclerosing cholangitis

Protease	Peptide origin	Peptide substrate distribution [Avg. ion counts ± SD]		P
		CCA case group	PSC/other BBD control group
ADAMTS4	Bile	670.33 ± 2100.31	95.35 ± 239.85	0.014
CMA1		353.85 ± 903.46	108.50 ± 225.54	0.0018
KLK4		407.48 ± 1309.74	68.33 ± 168.92	0.0072
ADAMTS4	Urine	429.51 ± 522.50	251.30 ± 434.86	0.006
CASP1		390.00 ± 627.39	537.32 ± 544.00	0.0007
KLK6		144.39 ± 189.94	289.47 ± 243.20	< 0.0001