Table 2.

Validation performance

Phenotype	no genus partition	with genus partition	difference
Endospores	0.946	0.821	-0.125
Gram stain	0.966	0.905	-0.061
Motility	0.932	0.870	-0.062
Oxygen Requirement	0.973	0.949	-0.024
average	0.954	0.886	-0.068

Comparison of phenotype prediction performance for different validation sets according to phylogenetic proximity of organisms (see also section "Methods"). The first column indicates the phenotype category, the second and third column represent the prediction performance in terms of the harmonic mean of sensitivity and specificity for the validation set without and with using the genus partition, respectively. Values in the fourth column correspond to the difference of prediction performance when the genus-partitioned data set is used.