From immunity to community: A predictive trait space model for white plague disease on Caribbean reefs

Abstract Disease outbreaks have played important roles in structuring communities. Therefore, the ability of individuals within populations to resist disease and their corresponding immune and molecular‐scale traits may provide a basis for predicting community dynamics of ecosystems facing disease outbreaks. Here, we utilized coral reefs as a model ecosystem to explore trait‐based models for disease‐disturbed community dynamics. Our trait space model incorporated intraspecific variation of immune traits to predict coral community assemblages after a white plague (WP) disease outbreak. Immune traits were measured from a WP transmission experiment conducted on seven coral species in the United States Virgin Islands (USVI). Susceptibility ranks were derived from the experiment and published survey data on USVI reefs. We compared the predictive power of six curated sets of host immune traits: environmentally plastic host genes, lineage‐specific host genes, host genes associated with disease response, host genes associated with stress response, immune protein antioxidant activity, and a combined group. An ideal trait distribution of disease resistance was created using the disease‐exposed individuals of the most resistant species and acted as the disturbance filter in the model. Our disease trait space model integrated a hierarchical Bayesian model to predict the ranked relative abundances of community members after the disturbance filter (i.e., WP outbreak) was applied. Results of several deliberate parameterizations determined multiple, curated immune trait sets that predicted post‐outbreak reef community dynamics better than null models. Traits that varied among species, particularly immune protein activity traits, rather than traits that responded to a disease disturbance best predicted community structure. The successful implementation of our disease trait space model links molecular‐scale traits with ecosystem level processes and suggests that immunity is a major driver of community dynamics.