Table 4. This table contains the results of our feature selection procedure for miRs profiled in urine.

Rank	Top 1	Top 2	Top 3	Top 4	Top 5	Top 6	Top 7	Top 8	miR
T-test
1	60	90	97	100	100	100	100	100	miR222
2	7	33	50	60	77	77	80	83	miR16
3	7	23	37	60	67	90	90	93	miR34a
4	0	3	37	50	67	80	83	87	miR182
5	20	27	33	47	53	53	63	70	miR375
6	0	7	13	23	37	50	53	63	miR205
7	0	3	10	17	27	37	70	77	miR331
8	0	3	7	7	20	33	47	57	miR21
9	0	0	0	13	20	30	40	67	miR20a
10	0	0	3	7	10	23	33	43	miR106b
11	0	0	0	0	3	7	13	30	miR221
12	0	3	7	10	13	13	20	23	miR145
Centroid
1	37	90	97	100	100	100	100	100	miR222
2	50	67	83	83	93	97	97	100	miR16
3	13	30	43	57	70	77	93	93	miR205
4	0	7	33	73	90	97	97	100	miR20a
5	0	3	27	50	80	90	97	100	miR331
6	0	3	10	10	20	63	93	100	miR21
7	0	0	7	17	23	43	70	87	miR375
8	0	0	0	3	10	20	40	93	miR34a
9	0	0	0	0	3	3	3	13	miR106b
10	0	0	0	7	10	10	10	13	miR218
11	37	90	97	100	100	100	100	100	miR222
12	50	67	83	83	93	97	97	100	miR16

The miRs have been ranked based on their appearance in the Top n miRs for different classifiers as part of the cross-validation procedure. The percentage of times the miR is selected by the classifier in the top n = 1..8 miRs is reported for both t-test and centroid approaches.