A machine learning approach for genome-wide prediction of morbid and druggable human genes based on systems-level data

Table 1 Classifier performance measures for prediction of morbid and druggable genes

Prediction of morbid genes
Performance measure	Median [min,max] ¹	Median [min,max] ¹	N	W	W_c (two-tailed p = 0.05)²
	Normal	Shuffled
Precision	0.658 [0.648,0.679]	0.495 [0.473,0.522]	10	0	8 *
Recall	0.648 [0.632,0.657]	0.502 [0.471,0.521]	10	0	8 *
AUC	0.716 [0.706,0.729]	0.498 [0.462,0.526]	10	0	8 *
Prediction of druggable genes
Performance measure	Median [min,max] ¹	Median [min,max] ¹	N	W	W_c (two-tailed p = 0.05)²
	Normal	Shuffled
Precision	0.748 [0.72,0.763]	0.5 [0.451,0.556]	10	0	8 *
Recall	0.782 [0.732,0.809]	0.492 [0.447,0.564]	10	0	8 *
AUC	0.820 [0.801,0.835]	0.500 [0.43,0.546]	10	0	8 *

ISSN: 1471-2164