Fig. 1

Co-occurrence gene network map showing positive (green lines) and negative gene correlations (red lines) for genes assigned to families. The size of the node denotes family relative abundance. Nodes in purple indicate putative keystone families and the gradient in colours relates the ranking of the families as ‘keystone’, where the darkest were the top ranked. The top cluster relates to primary colonisation (< 4 h) and the bottom secondary colonisation (> 4 h) events. Names of dominant and keystone familes are shown only due to the complexity of the network, for information on all interactions refer to Supplementary Tables 2 and 3