Efficacy of different protein descriptors in predicting protein functional families

Ong, Serene AK; Lin, Hong Huang; Chen, Yu Zong; Li, Ze Rong; Cao, Zhiwei

doi:10.1186/1471-2105-8-300

BMC Bioinformatics

Table 3 Dataset training statistics and prediction accuracies of six protein functional families. DS refers to descriptor set, where D1 = amino acid composition; D2 = dipeptide composition; D3 = Moreau-Broto autocorrelation; D4 = Moran autocorrelation; D5 = Geary autocorrelation; D6 = composition, transition and distribution descriptors; D7 = quasi sequence order; D8 = pseudo amino acid composition; D9 = combination of D1+D2; and D10 = combination of D1-D8. Predicted results given as TP (true positive), FN (false negative), TN (true negative), FP (false positive), Sen (sensitivity), Spec (specificity), Q (overall accuracy) and MCC (Matthews correlation coefficient).

From: Efficacy of different protein descriptors in predicting protein functional families

Protein family	Des-criptor set	Training set		Testing set				Independent evaluation set
		P	N	P		N		P			N			Q(%)	MCC
				TP	FN	TN	FP	TP	FN	Sen(%)	TN	FP	Spec(%)
EC2.4	D1	1249	2120	1154	1	9065	12	724	176	80.4	3244	202	94.1	91.3	0.74
	D2	1319	2120	1080	5	8806	1	646	154	82.9	3349	97	97.2	94.1	0.80
	D3	1105	1756	1295	4	9166	5	768	132	85.3	3394	52	98.5	95.8	0.87
	D4	1239	2221	1161	4	8701	5	756	144	84.0	3365	81	97.7	94.8	0.84
	D5	1242	2223	1160	2	8690	14	753	147	83.6	3391	55	98.4	95.4	0.85
	D6	1214	2077	1145	45	8846	4	741	159	82.3	3383	63	98.2	94.9	0.84
	D7	1293	2624	1072	39	8295	8	696	204	77.3	3270	176	94.9	91.3	0.73
	D8	1226	3008	1177	1	7918	1	794	106	88.2	3387	59	98.3	96.2	0.88
	D9	1275	2747	1129	0	8177	3	782	118	86.9	3367	79	97.7	95.5	0.86
	D10	1228	3254	1176	0	7672	1	798	102	88.7	3397	49	98.6	96.5	0.89
GPCR	D1	1590	7458	1847	1	14166	3	505	17	96.7	6735	58	99.1	99.0	0.93
	D2	564	711	1728	3	14121	5	510	12	97.7	6737	56	99.2	99.1	0.93
	D3	1169	4628	1122	4	10208	1	507	15	97.1	6737	56	99.2	99.0	0.93
	D4	1257	4474	1037	1	10363	0	499	23	95.6	6745	48	99.3	99.0	0.93
	D5	1290	4724	997	8	10113	0	494	28	94.6	6734	59	99.1	98.8	0.91
	D6	757	2060	1536	2	12777	0	503	19	96.3	6742	51	99.2	99.0	0.93
	D7	812	2950	1482	1	11887	0	495	27	94.8	6696	97	98.6	98.3	0.88
	D8	653	2171	1644	0	12550	1	501	21	96.0	6769	24	99.7	99.4	0.95
	D9	1590	7458	693	12	7322	57	512	10	98.1	6735	58	99.1	99.1	0.93
	D10	672	2454	1625	0	12268	0	502	20	96.2	6757	36	99.5	99.2	0.94
TC8.A	D1	118	2858	49	0	13121	0	36	27	57.1	1843	2	99.9	98.5	0.73
	D2	116	1100	50	0	14824	0	41	22	65.1	1843	2	99.9	98.7	0.78
	D3	94	7962	53	0	14501	0	42	21	66.7	1842	3	98.6	98.7	0.78
	D4	94	7962	47	0	11250	0	37	26	58.7	1843	2	99.9	98.5	0.74
	D5	94	7962	47	0	11137	0	37	26	58.7	1843	2	99.9	98.5	0.74
	D6	94	7962	64	0	15283	0	44	19	69.8	1843	2	99.9	98.9	0.81
	D7	94	7962	59	0	15045	0	43	20	68.3	1843	2	99.9	98.9	0.80
	D8	103	943	63	0	14981	0	48	15	76.2	1843	2	99.9	99.1	0.85
	D9	114	810	52	0	15114	0	41	22	65.1	1843	2	99.9	98.7	0.78
	D10	102	1068	64	0	14856	0	48	15	76.2	1843	2	99.9	99.1	0.85
Chlorophyll	D1	356	7928	166	0	14297	0	182	128	58.7	1587	11	99.3	92.7	0.71
	D2	4S40	934	248	1	7927	1	228	82	73.6	1595	3	99.8	95.6	0.83
	D3	425	603	264	0	15253	0	246	64	79.4	1594	4	99.8	96.4	0.86
	D4	415	574	273	1	15282	0	247	65	79.7	1597	1	99.9	96.6	0.87
	D5	429	615	259	1	15240	1	233	77	75.2	1597	1	99.9	95.9	0.84
	D6	482	946	202	5	14910	0	205	105	66.1	1597	1	99.9	94.4	0.79
	D7	394	3337	210	85	12517	2	178	132	57.4	1597	1	99.9	93.0	0.73
	D8	371	1421	317	1	14435	0	255	55	82.3	1593	5	99.7	96.9	0.88
	D9	399	1273	289	1	14582	1	249	61	80.3	1591	7	99.6	96.4	0.86
	D10	381	1753	307	1	14102	1	251	59	81.0	1594	4	99.8	96.7	0.88
Lipid synthesis	D1	849	2026	705	3	8229	7	470	165	74.0	1218	57	95.5	88.4	0.73
	D2	927	2037	629	1	8225	0	512	123	80.6	1259	16	98.6	92.7	0.84
	D3	898	2968	659	0	7294	0	509	126	80.2	1271	4	99.7	93.2	0.84
	D4	968	3227	588	1	7035	0	493	142	77.6	1273	2	99.8	92.5	0.83
	D5	970	3280	586	1	6982	0	491	144	77.3	1260	15	98.8	91.7	0.81
	D6	874	2112	681	2	8149	1	525	110	82.7	1268	7	99.5	93.9	0.86
	D7	863	2415	692	2	7845	2	512	123	80.6	1271	4	99.7	93.4	0.85
	D8	907	1608	615	0	4488	0	498	137	78.4	1268	7	99.5	92.5	0.83
	D9	815	1613	740	2	8638	11	525	110	82.7	1248	27	97.9	92.8	0.84
	D10	865	1640	657	0	4456	0	531	104	83.6	1268	7	99.5	94.2	0.87
rRNA binding	D1	548	579	3390	6	9598	22	1824	87	95.5	3511	60	98.3	97.3	0.94
	D2	1133	1225	2811	0	8974	0	1844	67	96.5	3519	52	98.5	97.8	0.95
	D3	1126	1638	2816	2	8560	1	1812	99	94.8	3535	36	99.0	97.5	0.95
	D4	1337	1958	2697	0	8241	0	1783	128	93.3	3484	87	97.6	96.1	0.91
	D5	1372	1976	2572	0	8223	0	1784	127	93.4	3479	92	97.4	96.0	0.91
	D6	921	1208	2971	52	8991	0	1824	87	95.5	3541	30	99.2	97.9	0.95
	D7	878	2743	3040	26	7442	14	1808	103	97.9	3481	90	97.5	96.5	0.92
	D8	810	2245	3143	0	7954	0	1849	62	96.8	3541	30	99.2	98.3	0.96
	D9	810	972	3075	3	9182	2	1848	63	96.7	3526	45	98.7	98.0	0.96
	D10	900	2600	3044	0	7599	0	1858	53	97.2	3547	24	99.3	98.6	0.97

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com