Gene expression profiling in whole blood identifies distinct biological pathways associated with obesity

BMC Medical Genomics

Table 2 Classification of lean and obese subjects

	*True Positive Rate (Sensitivity)*		*False Positive Rate*		*True Negative Rate (Specificity)*		*False Negative Rate*
Classifier	Full	Filtered	Full	Filtered	Full	Filtered	Full	Filtered
Naïve Bayes	0.96 (0.09)	0.93 (0.08)	0.18 (0.16)	0.003 (0.02)	0.82 (0.17)	0.99 (0.02)	0.03 (0.09)	0.06 (0.08)
Logistic Regression	0.98 (0.04)	0.95 (0.07)	0.11 (0.16)	0.01 (0.03)	0.89 (0.16)	0.99 (0.03)	0.01 (0.04)	0.05 (0.08)
Random Forests	0.95 (0.09)	0.94 (0.11)	0.17 (0.17)	0.10 (0.14)	0.83 (0.17)	0.89 (0.14)	0.05 (0.09)	0.06 (0.11)
ZeroR	0.81 (0.39)	0.81 (0.39)	0.81 (0.39)	0.81 (0.39)	0.19 (0.39)	0.19 (0.39)	0.19 (0.39)	0.19 (0.39)

Comparison of classifier performance for predicting obese and lean phenotype with a full (183 genes) or filtered (11 gene) gene inputs

ISSN: 1755-8794