Table 3.

Performance of the models on their corresponding complete data sets

	H1N1			H3N2
Protein	Sensitivity	Specificity	MCC	Sensitivity	Specificity	MCC
HA	0.999	0.953	0.961	1	0.987	0.993
M1	1	0.881	0.934	0.994	1	0.971
M2	1	0.859	0.918	0.996	0.873	0.908
NA	1	0.907	0.95	1	0.908	0.95
NP	1	0.864	0.92	0.994	0.957	0.946
NS1	0.998	0.932	0.954	0.991	0.993	0.96
NEP	0.995	0.883	0.912	0.997	1	0.988
PA-X	0.901	1	0.856	1	1	1
PA	0.972	0.979	0.892	0.996	0.979	0.969
PB1-F2	0.91	0.987	0.884	0.999	0.778	0.861
PB1	0.993	0.93	0.923	1	0.879	0.932
PB2	0.989	0.984	0.935	0.996	0.985	0.972

Sensitivity is the ability to correctly predict human sequences and specificity is the ability to correctly predict avian sequences where 1 means perfect prediction and 0 means no correct prediction. Matthews correlation coefficient (MCC) value is a measure of how well the model performs overall where 1 means a perfect classification, 0 is for a prediction no better than random and −1 indicates a total disagreement between predictions and observations. “na” means the measure could not be calculated for the given model