Table 6.

Pearson’s ${\chi }^2$ test $p$-values calculated between the predicted and ground truth distributions for WT NL4-3 and primary isolate YU2 and JR-CSF HIV-1 viruses. TEMNet is the only backbone that predicts a distribution with no statistical significance for both primary isolates within a probability threshold of $p=5\%$ when increasing the confidence threshold to 0.9. The highest average $p$-values are encountered for ResNet101v2, TEMNet, ResNet101 and Inception-ResNet-v2 in descending order. $p$-values greater than 0.05 indicate no significant statistical difference.

	${\chi }^2$ test $p$-value
Backbone	Confidence	NL4-3	NL4-3 delIN	YU2	JR-CSF
TEMNet	>0.5	0.18	0.03	0.76	0.02
	>0.9	0.08	0.01	0.06	0.28
ResNet101	>0.5	0.29	0.02	0.02	0.26
	>0.9	0.21	0.05	$2.03e^{-4}$	0.02
ResNet101v2	>0.5	0.58	0.46	$5.37e^{-4}$	$1.30e^{-4}$
	>0.9	0.51	0.54	$3.69e^{-5}$	$2.78e^{-3}$
Inception-ResNet-v2	>0.5	0.09	$1.90e^{-5}$	$1.32e^{-6}$	0.02
	>0.9	0.02	$2.57e^{-3}$	$2.91e^{-6}$	0.44