A unified frame of predicting side effects of drugs by using linear neighborhood similarity

BMC Systems Biology

Table 2 5-CV performances of prediction models on Liu’s dataset

Data	Methods	AUC	AUPR	Hamming Loss	Ranking Loss	One Error	Coverage	Average Precision
Enzyme	LNSM	0.8898	0.4187	0.0473	0.0821	0.1659	846.3846	0.4696
Pathway	LNSM	0.8886	0.4273	0.0470	0.0776	0.1647	814.6298	0.4932
Target	LNSM	0.8991	0.4708	0.0452	0.0690	0.1538	792.3726	0.5216
Transporter	LNSM	0.8896	0.4147	0.0477	0.0817	0.1611	849.3161	0.4762
Treatment	LNSM	0.9013	0.4836	0.0446	0.0710	0.1262	806.8558	0.5232
Substructure	LNSM	0.8944	0.4538	0.0459	0.0714	0.1490	803.5228	0.5184
All data	LNSM-SMI	0.8986	0.5053	0.0435	0.0670	0.1154	789.8486	0.5476

ISSN: 1752-0509