Two sets of histograms showing the sensitivity () and edge balance for the networks in which long limit cycles are suppressed.

2012-05-07T00:31:47Z (GMT) by Jamie X. Luo Matthew S. Turner

The 1000 networks of 5 nodes were evolved to maximise sensitivity to edge loses, with a fitness discount factor against long limit cycles. Panel A shows that while most networks are able to increase their sensitivity () to edge loses they are unable to achieve optimal sensitivity without the longer limit cycles. Panel B shows that, as under the previous evolutionary criteria the networks become more topologically balanced in order to increase their sensitivity. indicates the number of networks.