mGOASVM: Multi-label protein subcellular localization based on gene ontology and support vector machines

BMC Bioinformatics

Table 6 Distribution of the number of labels predicted by mGOASVM for proteins in the virus and plant datasets

Dataset	Condition	Case	$n_{k}^{o}$ , $n_{k}^{e}$ or $n_{k}^{u}$				(N^o, N^e or N^u)/N_act
			k =0	k =1	k =2	k >2
Virus	$\| ℳ (p_{i}) \| > \| ℒ (p_{i}) \|$	Over-prediction	0	18	0	0	18/207 = 8.7%
	$\| ℳ (p_{i}) \| = \| ℒ (p_{i}) \|$	Equal-prediction	187	0	0	0	187/207 = 90.3%
	$\| ℳ (p_{i}) \| < \| ℒ (p_{i}) \|$	Under-prediction	0	2	0	0	2/207 = 1.0%
Plant	$\| ℳ (p_{i}) \| > \| ℒ (p_{i}) \|$	Over-prediction	0	83	2	0	85/978 = 8.7%
	$\| ℳ (p_{i}) \| = \| ℒ (p_{i}) \|$	Equal-prediction	879	0	0	0	879/978 = 89.9%
	$\| ℳ (p_{i}) \| < \| ℒ (p_{i}) \|$	Under-prediction	0	14	0	0	14/978 = 1.4%

$|ℳ (p_{i})|$ : Number of predicted labels for the i-th (i=1,…, N_act) protein; $|ℒ (p_{i})|$ : Number of the true labels for the i-th protein; Over-prediction: the number of predicted labels is larger than that of the true labels; Equal-prediction: the number of predicted labels is equal to that of the true labels; Under-prediction: the number of predicted labels is smaller than that of the true labels; $n_{k}^{o}, n_{k}^{e}$ or $n_{k}^{u}$ : the number of proteins that are over-, equal-, or under-predicted by k (k=0,…,5 for the virus dataset and k= 0,…,11 for the plant dataset) labels, respectively; N^o, N^e or N^u: the total number of proteins that are over-, equal-, or under-predicted, respectively.

ISSN: 1471-2105