Skip to main content
. 2021 Mar 9;4:313. doi: 10.1038/s42003-021-01826-7

Fig. 7. Perspectives on host generalism, calculated with two metrics.

Fig. 7

a The average number of observed plant species with which OTUs affiliate in each plot was calculated from occurrence data and calculated as resource range index (RRI, see “Methods”; higher RRI represents a lower number of hosts per OTU). Endophyte richness was negatively associated with RRI (i.e., in plots with higher endophyte richness, the number of hosts used per OTU was higher) (R2 = 0.55, p = 0.004). b The proportion of shared OTUs was calculated as the proportion of OTUs that occurred in more than one plant species, divided by the total number of OTUs in the plot. Consistent with a, we found that as richness increased, the number of nonspecialists increased (R2 = 0.47, p = 0.009). We then considered the relationship of the RRI metric to seasonality, finding that (c) temperature seasonality and (d) precipitation seasonality could explain variation in RRI (R2 = 0.47, p = 0.009; R2 = 0.20, p = 0.08, respectively). Error bar represents standard errors. Dotted lines represent 95% confidence interval.