Sequence-based information-theoretic features for gene essentiality prediction

BMC Bioinformatics

Table 3 Leave-one-species-out results using SVM and Random Forest classifiers

	Our method		Liu et al.	Palaniappan and Mukherjee	Geptop (homology)	Geptop* (Composition)
Training on (No. of species)	14		30	14	18	18
	Random Forest	SVM	SVM	SVM	Score based	Score based
AB	0.81	0.83	0.75	0.74	0.85	0.79
BS	0.84	0.84	0.77	0.58	0.95	0.81
EC	0.87	0.88	0.83	0.65	0.95	0.84
FN	0.83	0.83	0.67	0.66	0.84	0.74
HI	0.75	0.77	0.54	0.46	0.57	0.59
HP	0.75	0.74	0.52	0.59	0.60	0.64
MG	0.68	0.66	0.60	0.64	0.72	0.56
MP	0.75	0.74	0.64	0.61	0.87	0.76
MT	0.80	0.77	0.70	0.49	0.73	0.77
PA	0.80	0.80	0.65	0.66	0.80	0.79
SA	0.88	0.90	0.81	0.66	0.84	0.86
SA2	0.86	0.85	0.80	-	0.88	0.83
SE	0.86	0.86	0.69	-	0.95	0.86
ST	0.81	0.79	0.84	0.53	0.71	0.69
VC	0.75	0.72	0.69	-	0.61	0.72
Average	0.80	0.80	0.70	0.61	0.79	0.75

The average AUC scores of four existing methods are also presented for comparison. Geptop* is a sequence composition based predictor presented along with Geptop [23]

ISSN: 1471-2105