Table 2. Performance comparison at threshold equal to 0.5 between MTDNN and other prediction methods.

This table reports average F₁, Precision, Recall scores of five methods in three GO domains, namely BP, MF, and CC, with the threshold equal to 0.5. MLDNN* is the performance obtained at the thresholds that maximise the F₁ scores for individual GO terms in the training set. The best F₁ scores in respective domains are marked in bold font.

Methods	BP			MF			CC
	Precision	Recall	F1	Precision	Recall	F1	Precision	Recall	F1
MTDNN	0.282	0.312	0.296	0.283	0.303	0.292	0.389	0.626	0.48
MLDNN	0.353	0.103	0.160	0.335	0.170	0.267	0.402	0.308	0.349
STDNN	0.070	0.676	0.127	0.112	0.606	0.189	0.168	0.858	0.281
FFPred	0.161	0.492	0.242	0.141	0.558	0.225	0.248	0.820	0.380
BLAST	0.129	0.019	0.033	0.447	0.035	0.066	0.423	0.027	0.051
Naive	0.539	0.024	0.046	0.318	0.117	0.171	0.446	0.407	0.425