A top-down approach to classify enzyme functional classes and sub-classes using random forest

EURASIP Journal on Bioinformatics and Systems Biology

Table 6 Performance of random forest classifier using Model 2, direct sub-class classification approach

Class label	Instances	True positive	Precision (%)	Recall (%)
1.1	192	157	83.51	81.77
1.10	88	87	93.55	98.86
1.16	74	65	89.04	87.84
1.2	179	169	88.02	94.41
1.3	176	162	94.74	92.05
1.4	175	162	94.74	92.57
1.5	97	78	81.25	80.41
2.1	96	80	95.24	83.33
2.2	91	90	70.87	98.9
2.3	80	39	60.94	48.75
2.4	98	83	91.21	84.69
2.5	97	86	88.66	88.66
2.6	96	82	82.83	85.42
2.7	110	68	93.15	61.82
2.8	93	74	94.87	79.57
3.1	87	71	77.17	81.61
3.11	36	33	80.49	91.67
3.2	89	85	86.73	95.51
3.3	87	76	87.36	87.36
3.4	95	86	94.51	90.53
3.5	91	85	91.4	93.41
3.6	92	88	92.63	95.65
3.7	55	43	91.49	78.18
3.8	36	31	79.49	86.11
4.1	172	159	90.34	92.44
4.2	185	147	89.63	79.46
4.3	183	180	91.84	98.36
4.4	91	77	91.67	84.62
4.6	97	85	91.4	87.63
4.99	93	87	92.55	93.55
5.1	179	168	85.28	93.85
5.2	96	86	83.5	89.58
5.3	188	154	84.15	81.91
5.4	98	77	92.77	78.57
5.5	99	84	85.71	84.85
6.1	418	416	93.91	99.52
6.2	183	156	81.68	85.25
6.3	170	151	87.79	88.82
6.4	82	70	80.46	85.37
Overall	4748	4190	87.35	86.74