Skip to main content
2023 ESA Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.
Back

Contributed Talk

Session: : Communities: Traits And Functional Diversity 6

COS 257-2 - From immunity to community: a predictive trait-space model for a Caribbean coral disease

Thursday, August 10, 2023
10:15 AM – 10:30 AM PDT
Location: C120-121
Abstract: Coral disease outbreaks have played important roles in structuring reef communities. Therefore, the ability of individuals within populations to resist disease and their corresponding immune traits may provide a basis for predicting community dynamics on reefs disturbed by disease. We developed a trait-space model that incorporates intraspecific variation of immune traits to predict coral community assemblages after a white plague disease outbreak. Using results from a white plague transmission experiment conducted in the United States Virgin Islands, seven species were identified as susceptible (Orbicella favelota and O. annularis), tolerant (Colpohylia natans and Siderastrea siderea), or resistant (Montastraea cavernosa, Porites porites, and P. asteroides). Traits considered in the model consisted of immune data (e.g. gene expression and protein activity data) gathered during the experiment. Multiple trait sets were used to compare the predictive power of different types of immune traits, including those related to temperature stress or other diseases. Influential traits for white plague resistance were determined by random forest analysis and Spearman correlations were used to reduce the number of redundant traits. An ideal trait distribution of disease resistance was created using the values of the disease-exposed individuals of the most resistant species, M. cavernosa. Mclust, an R package for normal mixture modeling, was used to compute the probability of a trait given for each species using the probability density functions of the actual traits present in the seven species. Traits were randomly drawn from the ideal traits distribution to calculate probabilities of a trait given the disturbance, where this disturbance filter was essentially the target trait distribution for being disease resistant. Our trait-space approach then integrated a hierarchical Bayesian model to predict the community assemblage after the disturbance filter (i.e. a white plague outbreak) was applied. Ranked relative abundances predicted by the model using the resistance levels derived from experimental data were tested against those of a null model that randomized traits within species and an alternative model that considered P. astreoides the most resistant species based on field survey data from the region. Results show the trait-space model predicts an accurate post-outbreak community structure when using immune traits that influenced white plague resistance. Additionally, the model determined that the antioxidant, catalase, and the immune protein, prophenyloxidase, were important immune traits for predicting community structure. Our trait space model suggests that immune traits alone are predictive drivers of Caribbean reef community assemblages subjected to disease outbreaks.