A protein network refinement method based on module discovery and biological information

BMC Bioinformatics

Table 8 The variation of the effect of thresholds on the selection of critical modules and the performance of the network

th₁	th₂	th₃	Number of critical modules	Top 100, 600,1130	ACC	PRAUC
0.015	1.5	0.25	15	84, 398, 563	0.761	0.564
0.015	2	0.25	13	84, 398, 551	0.756	0.569
0.015	1.5	0.5	13	85, 387, 532	0.748	0.555
0.015	2	0.5	15	83, 383, 533	0.748	0.550
−0.005	1.5	0.25	17	85, 396, 567	0.763	0.567
−0.005	2	0.25	15	85, 405, 567	0.763	0.572
−0.005	1.5	0.5	15	84, 395, 541	0.752	0.559
−0.005	2	0.5	17	86, 387, 541	0. 752	0.554
−0.02	1.5	0.25	20	84, 384, 556	0.758	0.554
−0.02	2	0.25	18	85, 387, 555	0.758	0.558
−0.02	1.5	0.5	17	85, 382, 531	0.748	0.551
−0.02	2	0.5	19	86, 383, 535	0.749	0.547

ISSN: 1471-2105