AnnoPRO: a strategy for protein function annotation based on multi-scale protein representation and a hybrid deep learning of dual-path encoding

Table 4 A comparison among the predictive performances of AnnoPRO and three representative methods for the functional annotations of two well-known growth differentiation factors (GDF8, GDF11)

Protein Name	Methods	BP		CC		MF
Protein Name	Methods	Recall	Precision	Recall	Precision	Recall	Precision
GDF8	DeepGOPlus	0.578	0.320	0.333	1.000	0.389	0.333
	PFmulDL	0.333	0.198	0.667	0.400	0.444	0.444
	NetGO3	0.351	0.806	1.000	0.375	1.000	0.783
	*AnnoPRO*	0.898	0.898	1.000	0.731	1.000	1.000
GDF11	DeepGOPlus	0.402	0.306	0.625	0.714	0.222	1.000
	PFmulDL	0.404	0.494	0.875	0.412	0.556	0.833
	NetGO3	0.553	0.547	0.750	0.750	0.778	1.000
	*AnnoPRO*	0.621	0.952	1.000	0.833	1.000	1.000

Those values indicating the best performances among all methods were highlighted in BOLD, and AnnoPRO performed the best in the vast-majority (11/12) of the Gene Ontology (GO) classes (BP, CC, MF) under both evaluating criteria (recall, precision). All methods were ordered based on their publication dates. BP Biological process, CC Cellular component, MF Molecular function, GDF8 Growth differentiation factor 8, GDF11 Growth differentiation factor 11

ISSN: 1474-760X