Collection

Complexity Research in Nature Communications

Ecology and evolution

The fin to limb transition as the re-organisation of a Turing pattern
Koh Onimaru et al., Nat. Commun. 7 11582 (2016)

Alan Turing suggested that morphological development could be regulated by a self-organizing, reaction-diffusion mechanism. Indeed, mouse digit patterning has been found to be controlled by a Turing network of Bmp, Sox9, and Wnt. Here, Onimaru and colleagues show that fin patterning in the catshark, Scyliorhinus canicula, is controlled by the same network with a different spatial organization. Thus, they demonstrate that the Turing network is deeply conserved in limb development.
 

Predicting the stability of large structured food webs
Stefano Allesina et al., Nat. Commun. 6 7842 (2015)

Random matrix theory has been central to the study of the relationship between complexity and stability in ecological networks. However, incorporating variable ecological interaction strengths and other dimensions of ecological complexity into these models has been a key limitation. Here, Allesina and colleagues use the ‘cascade model’, which does not assume a random graph, to develop model food web networks and analytically approximate the stability under various conditions. They find that network stability is determined by variability in interaction strengths, rather than the mean.
 

Social inheritance can explain the structure of animal social networks
Amiyaal Ilany and Erol Akçay, Nat. Commun. 7 12084 (2016)

Social network position influences many aspects of life not only for humans but also for other animals. Here, Ilany and Akçay develop a social network model in which social contacts can be inherited by offspring. They find that including such social inheritance produces networks with similar degree distribution, modularity, and clustering coefficient distribution as observed in the spotted hyena, rock hyrax, bottlenose dolphin, and sleepy lizard.


Environmental change makes robust ecological networks fragile
Giovanni Strona and Kevin D. Lafferty, Nat. Commun. 7 12462 (2016)

Despite their complexity, ecological networks appear robust to species loss. Here, Strona and Lafferty explore whether that robustness extends to novel environmental conditions. They simulate the dynamics of artificial and real-world host-parasite communities and show that secondary extinctions following the removal of a host species are limited under historical environmental conditions. However, they demonstrate that the same communities are likely to collapse when the environment changes.
 

Ecological networks are more sensitive to plant than to animal extinction under climate change
Matthias Schleuning et al., Nat. Commun. 7 13965 (2016)

Ecological networks involve numerous interdependencies, such as between plants and their pollinators, and the ability of a species to persist despite changing environmental conditions could depend on the strength and number of its interactions. Here, Schleuning and colleagues couple simulation models with empirical networks (plants and either their pollinators or seed-dispersers). They show that animal persistence under climate change depends more strongly on plant persistence than vice versa.
 

No complexity–stability relationship in empirical ecosystems
Claire Jacquet et al., Nat. Commun. 7 12573 (2016)

Most studies of the complexity-stability relationship in ecological networks have used randomly-built in silico networks and have supported initial theoretical predictions that more complex ecological networks are, counterintuitively, less stable. However, Jacquet and colleagues demonstrate that this prediction does not stand in a stability analysis of over 100 empirical food webs. This work highlights the need for network theories to better account for the non-random properties of ecological networks.