Predicting primate-parasite associations using exponential random graph models

Area of Science:

• Ecology
• Parasitology
• Network Analysis

Background:

• Host-parasite interactions are shaped by host exposure, susceptibility, and parasite transmission modes.
• Network analysis offers advanced tools to study ecological networks and host-parasite relationships.

Purpose of the Study:

• To investigate the biogeographic, phylogenetic, and ecological traits influencing interactions between nonhuman primates and their parasites.
• To test existing hypotheses and explore new predictors of host-parasite associations using a comprehensive database.

Main Methods:

• Utilized a network-based framework, specifically exponential random graph models (ERGMs).
• Analyzed an expanded host-parasite database with 213 primate nodes, 763 parasite nodes, and 2319 interactions.
• Incorporated analyses of phylogenetic relatedness, sampling effort, and spatial overlap among hosts.

Main Results:

• More threatened hosts exhibited fewer parasites, independent of their geographic range.
• Threatened species shared more parasites, suggesting a loss of specialist parasites and reliance on generalist ones.
• Viruses, protozoa, and helminths displayed broader host ranges than bacteria or fungi.
• Parasites infecting non-primates were more likely to infect multiple primate species.

Conclusions:

• The ERGM approach provides a comprehensive view of factors structuring host-parasite networks.
• Identified new associations between host and parasite traits, highlighting the importance of network structure in host-parasite dynamics.
• Results support previous findings and reveal novel insights into primate-parasite ecological networks.