An improved classification of G-protein-coupled receptors using sequence-derived features

BMC Bioinformatics

Table 1 The number of proteins in four datasets and the corresponding prediction accuracies.

Dataset	Family/sub-subfamily	Tot(i)	c(i)	ACC(%)
D167	Acetylcholine	31	31	100
	Adrenoceptor	44	44	100
	Dopamine	38	36	94.74
	Serotonin	54	53	98.15
	Overall	167	164	98.2
D566	Adrenoceptor	66	65	98.48
	Chemokine	92	90	97.83
	Dopamine	43	40	93.02
	Neuropeptide	31	30	96.77
	Olfactory	84	84	100
	Rhodopsin	183	180	98.36
	Serotonin	67	65	97.01
	Overall	566	554	97.88
D1238	Rhodopsin-like	1103	1102	99.91
	Secretin-like	84	83	98.81
	Metabotrophic/glutamate/pheromone	51	50	98.04
	Overall	1238	1235	99.76
D365	Rhodopsin-like	232	222	95.69
	Secretin-like	39	34	87.18
	Metabotrophic/glutamate/pheromone	44	39	88.64
	Fungal pheromone	23	22	95.65
	CAMP receptor	10	10	100
	Frizzled/smoothened	17	11	64.71
	Overall	365	338	92.6

Tot(i) is the number of sequences observed in class i, c(i) is the number of correctly predicted sequences of class i, and ACC is the prediction accuracy.

ISSN: 1471-2105