Identifying the drivers for disease prevalence across multiple scales and hierarchies is a fundamental step for disease ecology to advance from a one-to-one framework to a multi-host multi-parasite predictive framework. Various approaches have been developed to tackle this level of complexity in order to understand and better predict disease outcomes in ecological systems. We understand that species don’t exist in isolation and that parasite transmission is influenced by host-parasite compatibility. In the case of avian malaria, an intricate network of parasite and host relationships, along with vector characteristics and environmental conditions, play a role determining disease transmission. The disparity between extreme levels of disease prevalence among different species of hosts has motivated research on parasite and host community assembly, and the different drivers of disease. Taking a metacommunity approach to a tropical avian malaria system, a set of different interaction scenarios were tested in order to explore the influence of environmental variables and host phylogenetic relatedness on parasite community assemblages and overall disease prevalence. A variation of Hierarchical Modeling for Species Communities (HMSC), a Bayesian joint species distribution modeling framework, is used to analyze the avian malaria data to understand processes determining parasite community assemblages at the host species level.
Results/Conclusions
Given the complexity of muti-host multi-parasite systems, different levels of interactions are at play when estimating disease prevalence and the specific parasite assemblages associated to host species. A metacommunity approach from the parasite perspective, in which different patches are analogous to host species, shows to be a useful framework to analyze disease community data. This approach allowed to model environmental variables from the parasite’s perspective as characteristics associated to the host and dispersal components associated to the disease vector. Implementing HMSC in this scenario allowed to identify the host phylogenetic relatedness as one of the strongest drivers for parasite community assemblage. It is important to note that this modeling approach has only considered presence or absence of parasites, not the parasite load, and did not consider disease progression as having an impact on the quality of the environment provided by the host. This work shows that a metacommunity approach is a useful way to analyze complex hierarchical data in a disease ecology scenario. It also shows that our way of analyzing the variation explained by the HMSC model can provide useful insight into the drivers of disease prevalence for complext communities