Detecting phylogenetic signal in mutualistic interaction networks using a Markov process model.

Area of Science:

• Ecology
• Evolutionary Biology
• Network Science

Background:

• Ecological interaction networks (e.g., plant-pollinator, plant-frugivore) display complex, non-random structures.
• Existing models fail to fully explain the formation and structure of these networks.
• The evolutionary history of interacting species is hypothesized to be a key factor shaping network architecture.

Purpose of the Study:

• To develop a generative model explaining how the evolution of individual species influences the evolution of ecological interaction networks.
• To test the hypothesis that phylogenetic history is a significant driver of interaction network structure.

Main Methods:

• A novel model was developed to describe the evolution of pairwise interactions using a branching Markov process.
• The model draws upon established phylogenetic models from molecular evolution.
• The model's predictions were compared against empirical data from plant-pollinator and plant-frugivore mutualistic networks.

Main Results:

• The proposed model, incorporating phylogenetic information, provided a significantly better fit to 21% of the studied ecological networks.
• This suggests that the inheritance of interaction patterns through evolutionary history plays a crucial role in network formation.
• The findings do not exclude the role of ecological novelties in shaping current network architectures.

Conclusions:

• Evolutionary history is a significant factor in shaping ecological interaction networks.
• The developed model offers a valuable tool for understanding the evolutionary basis of network structure.
• The model can serve as a null model to disentangle the effects of evolutionary history from other factors influencing network emergence.