Protein-protein binding site identification by enumerating the configurations

BMC Bioinformatics

Table 10 Detailed Results for DoBi on the data set used by core-SVM

Protein ID	Partner ID	*Int* _n ^a	C _n ^b	P _n ^c	Acc^d	Cov^e
1a9xA	1a9xB	59	52	95	54.7	88.1
1a9xB	1a9xA	52	47	88	53.4	90.4
1aym1	1aym3	46	38	41	92.7	82.6
1aym2	1aym1	57	54	70	77.1	94.7
1aym3	1aym1	43	33	36	91.7	76.7
1blxA	1blxB	21	15	33	45.5	71.4
1fzcB	1fzcC	45	38	58	65.5	84.4
1g4yR	1g4yB	29	5	18	27.8	17.2
1gk8A	1gk8I	49	28	55	50.9	57.1
1h1rB	1h1rA	33	9	14	64.3	27.2
1h8eC	1h8eD	69	37	67	55.2	53.6
1h8eD	1h8eC	35	19	39	48.7	54.3
1hxs4	1gxs1	31	21	35	60.0	67.7
1irdB	1irdA	23	20	32	62.5	86.9
1j34A	1j34B	43	19	22	86.4	44.1
1jboB	1jboA	36	16	29	55.2	44.4
1jsdA	1jsdB	51	18	20	90.0	35.3
1jsdB	1jsdA	67	26	42	61.9	38.8
1k5nA	1k5nB	35	24	56	42.9	68.6
1k5nB	1k5nA	25	16	39	41.0	64.0
1ld8A	1ld8B	35	23	28	82.1	65.7
1mtyB	1mtyD	58	22	34	64.7	38.1
1mtyD	1mtyB	31	10	15	66.7	32.2
1mtyG	1mtyD	41	18	42	42.9	43.9
1n4qB	1n4qA	25	5	15	33.3	20.0
1p2jA	1p2jI	23	18	36	50.0	78.2
1p2jI	1p2jA	14	13	21	61.9	92.9
1qopA	1qopB	35	32	52	61.5	91.4
1qopB	1qopA	34	31	51	60.8	91.2
1rthA	1rthB	57	32	68	47.0	56.1
1rthB	1rthA	58	33	69	47.8	56.9
1rypB	1rypA	31	13	24	54.1	41.9
1rzhH	1rzhM	37	8	16	50.0	21.6
1rzhL	1rzhM	48	42	45	93.3	87.5
1rzhM	1rzhL	51	45	48	93.8	88.2
1s5dD	1s5dA	4	4	29	13.7	100
1tugA	1tugB	17	14	39	35.9	82.4
1tugB	1tugA	12	9	24	37.5	75.0
1tx4B	1tx4A	25	18	34	52.9	72.0
1uvqA	1uvqB	61	35	39	89.7	57.4
1uvqB	1uvqA	55	26	31	83.9	47.2
1we3F	1we3T	12	10	48	20.8	83.3
1wf4o	1wf4a	10	10	19	52.6	100
2ltnA	2ltnB	55	12	16	75.0	21.8
2ltnB	2ltnA	47	17	17	100	36.2
3pcgA	3pcgM	41	12	15	80.0	29.3
3pcgM	3pcgA	40	11	21	52.4	27.5
4ubpA	4ubpC	24	8	43	18.6	33.3
4ubpC	4ubpB	46	26	86	30.2	56.5
8rucI	8rucA	38	29	38	76.3	76.3

^aInt_n is the number of residues on actual interface in complex.
^bC_n is the number of residues correctly predicted to be on interface by our method.
^cP_n is the number of total residues predicted to be on interface by our method.
^dAcc (%) is the accuracy of our method on the data set.
^eCov (%) is the coverage of our method on the data set.

ISSN: 1471-2105