SCMBYK: prediction and characterization of bacterial tyrosine-kinases based on propensity scores of dipeptides

BMC Bioinformatics

Table 4 The performance of 10 independent runs using BYK-TRN1102

	Fitness	Training ACC (%)	Test ACC (%)	MCC	Sen.	Spe.	Threshold
#1	99.21	97.36	96.27	0.93	0.97	0.95	474
#2	99.24	97.55	96.55	0.93	0.97	0.96	475
#3	99.21	97.82	96.36	0.93	0.97	0.95	486
#4	99.08	97.73	96.82	0.94	0.98	0.95	485
#5	99.32	97.64	96.82	0.94	0.99	0.95	484
#6	99.16	97.36	96.00	0.92	0.99	0.93	460
#7	99.02	97.00	96.27	0.93	0.95	0.98	496
#8	98.94	97.18	96.18	0.92	0.98	0.95	470
#9	99.08	97.82	96.91	0.94	0.97	0.97	464
#10	99.20	97.55	96.73	0.93	0.96	0.98	468
AVEG	99.19	97.50	96.49	0.93	0.97	0.96	476.20

ISSN: 1471-2105