BMC Bioinformatics

Table 2 Training sets used in the cost-sensitive experiment

From: Improving peptide-MHC class I binding prediction for unbalanced datasets

	A0203		A1101		A3101		B0702		B1501
%pos	B	NB	B	NB	B	NB	B	NB	B	NB
05	24	456	26	495	16	304	7	150	6	128
10	48	432	52	469	32	288	15	142	13	121
25	120	360	130	391	80	240	39	118	33	101
50	240	240	261	261	160	160	79	79	67	67
75	360	120	391	130	240	80	118	39	101	33
90	432	47	469	52	288	32	142	15	121	13
95	456	24	495	26	304	16	150	7	128	6

Number of binders (B) and nonbinders (NB) in training sets.

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com