An integrated approach to the detection of colorectal cancer utilizing proteomics and bioinformatics

Table 1 Descriptive data for the 4 potential biomarkers in the pattern discriminating CRC patients from HP (mean ± SD)

M/Z (Da)	AUC	HP	CRC patients
5 911	0.908	0.824 ± 0.504	2.763 ± 1.720
8 922	0.872	1.254 ± 0.724	2.767 ± 1.445
8 944	0.828	0.999 ± 0.626	2.651 ± 1.851
8 817	0.811	0.878 ± 0.607	1.744 ± 0.940

AUC: Area under the carve.

Table 2 Predicted results of classifier for discriminating CRC patients from HP

	Test set (49 × 10)		Training set (98 × 10)
	HP	CRC	HP	CRC
HP (92 × 10)	287	25	599	9
CRC (55 × 10)	19	159	29	343
Specificity (%)	92 (287/(287 + 25))		99 (599/(599 + 9))
Sensitivity (%)	89 (159/(159 + 19))		92 (343/(343 + 29))
Positive value (%)	86 (159/(159 + 25))		97 (343/(343 + 9))

Table 3 Descriptive statistics for the 7 potential biomarkers in the pattern discriminating CRC patients from CRA patients (mean ± SD)

M/Z (Da)	P value (× 10-5)	CRA patients	CRC patients
17 247	0.71	0.211 ± 0.130	0.113 ± 0.100
18 420	1.27	0.039 ± 0.036	0.076 ± 0.040
5 911	1.71	1.459 ± 0.977	2.763 ± 1.720
9 294	2.76	0.617 ± 0.385	1.105 ± 0.563
4 654	6.74	0.503 ± 0.493	1.164 ± 0.943
21 694	7.48	0.489 ± 0.145	0.698 ± 0.267
21 742	12.10	0.536 ± 0.161	0.744 ± 0.282

Table 4 Predicted results of classifier for discriminating CRC patients from CRA

	Test set (30 × 3)		Training set (60 × 3)
	CRA	CRC	CRA	CRC
CRA (35 × 3)	29	6	56	14
CRC (55 × 3)	6	49	16	94
Specificity (%)	83 (29/(29 + 6))		80 (56/(56 + 14))
Sensitivity (%)	89 (49/(49 + 6))		85 (94/(94 + 16))
Positive value (%)	89 (49/(49 + 6))		87 (94/(94 + 14))