Allesina & Tang (2012 Nature): An extension of analytic, Wigner semicircle theorem-like, stability criteria to predator-prey, competition and mutualism cases. Finds predator-prey interactions permit stability in networks as large and complex as real ones. Hierarchy is: competition-mutualism mixture, random, predator-prey. Finds that stability is less likely when both predator-prey networks and mutualistic networks are given more realistic structure (i.e. niche-model structure and nestedness respectively).