Measuring Nestedness: A comparative study of the performance of different metrics

Abstract: Nestedness is a property of interaction networks widely observed in natural mutualistic communities. Despite a widespread interest on this pattern, no general consensus exists on how to measure it. Instead, several metrics aiming at quantifying nestedness, based on different but not necessarily independent properties of the networks, coexist in the literature blurring the comparison between ecosystems. In this work, we present a detailed critical study of the behavior of six popular nestedness metrics and the variants of two of them. In order to evaluate their performance, we compare the obtained values of the nestedness of a large set of real networks among them and against a maximum entropy and maximum likelihood null model. Our results point out, first, that the metrics do not rank the degree of nestedness of networks universally. Furthermore, several metrics show significant undesired dependencies on the network properties considered. The study of these dependencies allows us to understand some of the systematic shifts between the real values of nestedness and the average over the null model. This paper intends to provide readers with a critical guide on how to measure nestedness patterns, by explaining the functioning of six standard metrics and two of its variants, and then disclosing its qualities and flaws. By doing so, we also aim to extend the application of the recently proposed null models based on maximum entropy to the still largely unexplored area of ecological networks. Finally, to complement the guide, we provide a fully-documented repository named nullnest which gathers the codes to produce the null model and calculate the nestedness index -- both the real value and the null expectation -- using the studied metrics. The repository contains, moreover, the main results of the null model applied to a large dataset of more than 200 bipartite networks.