07-31, 12:50–13:00 (UTC), Purple Track
In this talk, we will present three novel and complementary Julia packages that can handle data on species interactions. These packages can be used to import, simulate, analyze, and visualize all sorts of ecological networks, greatly simplifying the study of this emerging subfield of biology.
Network ecology is the novel approach to studying biodiversity. In that respect, this presentation provides an overview of three brand-new packages (Mangal, EcologicalNetworks, and EcologicalNetworksPlots) written by Timothée Poisot and collaborators, which were created in response to the increasing need to handle species interaction data in Julia. Built around state-of-the-art practices in network ecology, they largely facilitate the analysis and visualization of almost all types of species interaction networks. In these, two species are linked together if they can interact.
The Mangal package is a wrapper around the Mangal database, which is among the world's most exhaustive open source database of ecological interactions. The Mangal database contains 172 well-documented datasets for a total of 1300 ecological networks worldwide. It encompasses various types of ecological networks, including predation (food webs), pollination and parasitism networks.
In addition, the EcologicalNetworksPlots package implements the visualization of ecological networks, whereas EcologicalNetworks provides functions for the analysis of their emergent structure. The structure of species interaction networks is a rich source of information, as it is associated with the stability, dynamics, and functions of biological communities. The EcologicalNetworks package is the only one designed to conduct such a specialized analysis in Julia. It is also a suitable tool for the simulation of networks under different sets of assumptions and ecological models. These two packages use a well-documented type system for networks in order to capture the various types of data typically handled by network ecologists. This makes them easily extensible and somewhat fast.
Together, these three packages provide an extremely valuable methodological framework for the analysis of ecological networks in Julia. Our presentation offers an overview of their key features and depicts some of their concrete applications in this complex and emerging subfield of biology. It is aimed at ecologists among the Julia community, but also more broadly at network scientists and other attendees fascinated by networks and graphs.
PhD student in biological sciences in the computational ecology lab of Timothée Poisot (University of Montreal) and the integrative ecology lab of Dominique Gravel (University of Sherbrooke).