###Input files of PhyloH qiime_mapping_file.txt The mapping file generated for all data used in this case of study. This file contains all of the information about the samples necessary to perform the data analysis. The file is checked by QIIME script: check_id_map.py. seqs_otu.txt The file is generated by QIIME script: pick_otus.py. Produced by Usearch within Qiime the file define OTU, first item in each line followed by read in the different sample. The read name is simply sampleID_readID rep_phylo.tre Tree file in newick format ending with a semicolumn. Name of taxa need to match OTU name of seqOTU file. Branch length need to be specified Branch label, if present, will be overwritten by PhyloH to define univoque label The tree is generated by QIIME script: make_phylogeny.py. ###Output files of PhyloH tree.html PhyloH output in html format for "Status" variable. The file include partitioning information table results and pacman hairy tree plot. The file can be open with a internet browser and explored using functions such as text finder and zoom. Three types of data are shown on tree. 1) The color of the branches cyan indicates a contribution to I(T|E) (Beta diversity or Mutual information) significantly different from the null distribution, while branches are black otherwise. 2) The background of each branch is a gradient (color range) from yellow to red for increased contribution to I(T|E). 3) Bar plot on each tips indicates the number of reads count in each group. To browse the image: Look at the tree, find an relevant branches and text search the label of the branches to access the correct row on the node statistics table. Per Node Statistics in tabular format. ##Column names: NodeName:Names of nodes or tips present on the tree. betaA: Contribution to Mutual information OTU and Environment Significant: logical value after sequential Bonferroni ag['group1, group2,...']: Frequency of descendents of that branch in the groups agCount: Counts of descendents of that branch in the groups betaCA: Contribution to Mutual information of OTU and Sample within Environment It is possible use ITol web tools (http://itol.embl.de/) to modify the tree as to add further data set (i.e. taxonomic name) tree.xml xml file of tree.html tree_label.TreeLabeled Phylogenetic tree used in PhyloH analysis in Newick tree format. ###Supplementary Figures and Tables Fig_S1 The figure depicts how Shannon entropy is totally blind to some meaningful biological pattern. The Shannon entropy (i.e. nats and D) of field 1 is identical to field 2, even if the field 2 was found an organism, a dragonfly (order Odonata), very different from all the sampled butterflies (order Lepidoptera). Taking into account the phylogenetic entropy fields 1 and 2 are different. Indeed, assuming that the 3 butterflies belong to Pieridae, Sphingidae, Zygaenidae families, the dragonfly belong to the family Corduliidae and using a tree taken from TimeTree (www.timetree.org) the field 1 would have a diversity of 1.62 equivalent lineage, while field 2 would have 2.35 equivalent lineage, describing accurately the major contribution to diversity of Odonata in this sample. All silhouettes were downloaded from http://phylopic.org/ and drawn by Gareth Monger. Fig_S2.pdf Distribution of OTUs of High Frequency Cluster (HFC) between parasitised honey bee and varroa considering relative abundance. Fig_S3.pdf Distribution of OTUs of High Frequency Cluster (HFC) between healthy and parasitised honey bee and varroa. Relative abundance data are split up by Cells, Localities and Status variables. Letters are refered to the seven Localities. Fig_S4.pdf UPMGA trees showing the different relationships among samples for all the distance metrics considered in DMA (jaccard, squared-chord, unweighted UniFrac and weighted UniFrac), considering Localities and Status variables. Fig_S5.pdf Output of the script compute_core_microbiome.py. The plot has been used in the core community analysis to define optimal threshold. On the x axis is represented the fraction of OTUs occurring in a specific percentage of the total samples; while on the y axis is given the total number of OTUs considered (i.e. High Frequency Cluster, HFC). Table_S1.xls Summary of all OTUs taxon assignment. Table S2.xls Summary of results of adonis statistic.