A machine learning strategy for predicting localization of post-translational modification sites in protein-protein interacting regions

Table 3 Resulting performance measures when only sub-datasets corresponding to indices in the optimal sets were used in the classification tasks

Dataset	F₁	S_n	S_p	PPV	ACC	AUC	MCC	#indices used (of 102 total)	#features used (of 1428 total)
Acetylation	0.87^a	0.82	0.93	0.93	0.87	0.90^a	0.76^a	71	994
Phosphorylation	0.89^a	0.81^a	1.00	0.99	0.90^a	0.92	0.82^a	31	434
Ubiquitylation	0.86	0.81	0.92	0.93^a	0.87	0.90	0.74^a	86	1204

The maximum possible value for all measures is 1
^aSignificantly increased (t-test, α < 0.05) when compared with the use of all 102 indices