Dear members of the Methods in Ecology and Evolution editorial board,
On behalf of my co-authors, I am pleased to submit our manuscript entitled “Predicting metawebs: transfer of graph embeddings can help alleviate spatial data deficiencies” for consideration as a Perspective Article in Methods in Ecology and Evolution. This manuscript is the off-shoot of another manuscript that is currently under review at Methods in Ecology and Evolution (ID MEE-21-11-855.R1) and discusses topics previously included in the original manuscript, that were removed so as to keep the focus of the original manuscript on the methodological framework and maintain some sense of brevity. We have discussed this decision to split the original manuscript with the EiC Prof. Nicolas Lecomte, who has approved this decision. This submission thus puts forward a broader discussion as to how the transferring of graph embeddings could prove to be a practical tool in addressing the spatial data deficiencies we are faced with when working with ecological networks, and provides an overview of the graph embedding techniques.
Metawebs (networks of potential interactions within a species pool), have the potential to be a powerful starting point from which to understand how large-scales species interaction networks are structured. However, the challenges inherit in sampling species interactions makes assembling sufficient data at a large enough scale a tedious and costly process. Predictive methods can help in circumventing these limitations, of these predictive tools available graph embeddings (rather than focusing on the list of species interactions) holds, great potential. In this perspective, we outline how the challenges associated with inferring metawebs line-up with the advantages of graph embeddings; as well as how the choice of the species pool not only has consequences on the reconstructed network, but also embeds hypotheses about which human-made boundaries are ecologically meaningful.
The work described here has not been published elsewhere. All authors agree with the the manuscript content and its submission to the journal. No part of this manuscript has been published in any form elsewhere or is being considered for publication elsewhere while it is being considered for publication in this journal. This manuscript appeared as a preprint on EcoEvoRXiV with the following DOI: 10.32942/osf.io/vyzgr.
We look forward to your editorial decision on this manuscript.
Suitability statement:
The discussion in this perspective piece will help in further the discussion on the use and applicability of metawebs as well providing readers with an overview of how the use of graph embedding can help us to construct ecologically plausible metawebs with which to work.