Shedding-weighted network approaches for understanding tuberculosis maintenance in multihost systems using camera traps.

Understanding the dynamics of multihost pathogens, such as Mycobacterium tuberculosis complex (MTC), requires considering not only host interaction patterns but also variation in infectiousness across species. Network analysis is a useful tool to assess contact structure and disease risk, but it often depends on invasive methods. Camera trapping offers a noninvasive alternative to build co-occurrence networks in complex communities. In this study, we applied a novel approach that integrates shedding into ecological networks, weighing links between species pairs according to their co-occurrence frequency and shedding capacity, to evaluate tuberculosis (TB) risk across 18 study sites in the Iberian Peninsula. At the community level, TB risk was positively associated with mean strength-out, a local centrality measure of the frequency of interspecific contacts and their infectious potential. Species-specific models revealed that community TB risk increased with the strength-out of wild boar, red deer, and cattle, and with the closeness of red fox and badger. Furthermore, the community TB risk was jointly explained by shedding-weighted connectivity and spatial aggregation of wild boar, whereas red deer mainly contributed through their local abundance. In contrast, shedding-weighted centrality of badgers, foxes, and cattle explained TB risk, suggesting that ecological and management factors may influence TB spread. We defined epidemiological scenarios according to latitude, population factors, and infection pressure. These findings highlight the importance of including host infectiousness in ecological network analyses, as well as combining centrality measures and population data to understand and manage TB risk in complex host communities, with potential applications to other wildlife diseases and multihost systems.