Fig. 3

Relative abundance of the main species in each sample. The profiles of samples are ordered by Ward clustering of the squared Bray-Curtis distances between samples. The cluster dendrogram represents the similarity between samples regarding its microbial composition. Variables information for each sample (from lower to upper level: Farm, Treatment, Consistency-dpw) are indicated in the coloured squares below the bars. Numbers besides each branch refers to the main groups defined by Ward clustering, for which indicator species of each group determined by ‘multipatt’ are shown at the bottom part of the figure. Taxonomic identification of sequences was performed using Kaiju