Fig. 3: Drivers of viral taxonomic and functional diversity and compositions.
Path diagram for SEM showing only significant direct and indirect effects of biotic and abiotic variables on viral taxonomic a and functional b diversity and compositions. Composition is represented by the PC1 from the Bray-Curtis dissimilarity-based principal coordinate analyses. Numbers adjacent to the arrows are standardised path coefficients (r), analogous to relative regression weights and indicative of the effect size of the relationship. Blue and red arrows represent significant (P < 0.05) positive and negative pathways, respectively. Double-headed arrows indicate covariance between variables, single-headed arrows indicate a one way directed relationship. R2 represents the proportion of variance explained for every dependent variable in the model. The fit of models was evaluated using one-tailed Chi-squared test and root mean square error of approximation (RMSEA). vOTUs, the number of viral operational taxonomic units; PCs, the number of viral protein clusters; Ferric, ferric iron; MAT, mean annual temperature; MAP, mean annual precipitation; and MAGs, the number of metagenome-assembled genomes. Source data are provided in the Source Data file.
