Fig. 2 in Predicting the global mammalian viral sharing network using phylogeography

Fig. 2 The predicted viral sharing network predicts observed trends in an independent dataset. In all figures, points are jittered along the x-axis according to a density function; the black points and associated error bars are means ± standard errors. a Species pairs with higher predicted viral sharing probability from our model were more likely to be observed sharing a virus in the independent EID2 dataset. This comparison excludes species pairs that were also present in our training data. b Species that hosted a zoonotic virus in our dataset had more viral sharing links in the predicted all-mammal network than those without zoonotic viruses. c Species that had never been observed with a virus have fewer links in the predicted network than species that were known to host viruses in the EID2 dataset only, in our training data only, or in both. The y-axis represents viral sharing link number, scaled to have a mean of 0 and a standard deviation of 1 within each order for clarity. Black points represent means; error bars represent standard errors. Supplementary Figure 5 displays these same data without the within-order scaling.