Fig 1.
3-way edges and intersections.
(a) A small, 3-way network consisting of 5 nodes v1, v2, v3, v4 and v5 and two 3-way edges e1 and e2. Edge e1 connects nodes v3, v4 and v5 and edge e2 connects nodes v1, v2 and v3. (b) Venn diagram for a 3-way intersection of species. a is the number of families present in species A, b is the number of families present in species B, c is the number of families present in species C, ab is the number of families present in species A and species B, ac is the number of families present in species A and species C, bc is the number of families present in species B and species C, abc is the number of families present in species A, B and C, is the number of families present only in species A, is the number of families present only in species B and is the number of families present only in species C.
Fig 2.
Best-Edges 3-way Sørensen Network.
3-way Sørensen network pruned by selecting the best and second best edge for each node. Nodes represent bacterial species and edges represent similarity between triplets of bacterial species based on gene family content, quantified using the 3-way Sørensen Index. Nodes are coloured according to genus. Default colour is grey.
Fig 3.
(a) 2-way Sørensen Best Edges Network (b) Maximum Spanning Tree (MST) of the all-vs-all Sørensen network. Nodes represent bacterial species and edges represent similarity between pairs of bacterial species based on gene family content, quantified using the 3-way Sørensen Index. Nodes are coloured according to genus. The same node colour key as in Fig. 2 applies.
Fig 4.
Best-Edges 3-way Czekanowski Network.
3-way Czekanowski network pruned by selecting the best and second best edge for each node. Nodes represent bacterial species and edges represent similarity between triplets of bacterial species based on gene family content, quantified using the 3-way Czekanowski Index. Nodes are coloured according to genus. The same node colour key as in Fig. 2 applies.
Fig 5.
(a) 2-way Czekanowski Best Edges Network (b) Maximum Spanning Tree (MST) of the all-vs-all Czekanowski network. 3-way Sørensen network pruned by selecting the best and second best edge for each node. Nodes represent bacterial species and edges represent similarity between pairs of bacterial species based on gene family content, quantified using the 3-way Sørensen Index. Nodes are coloured according to genus. The same node colour key as in Fig. 2 applies.
Fig 6.
Network of bacteria species connected through shared enriched gene families. Small, white nodes represent gene families, coloured nodes represent bacterial species coloured by genus. Edges connect gene families to species in which they are enriched.
Fig 7.
Clostridium and Bacillus subnetwork.
Subnetworks containing the Clostridium and Bacillus species selected from (a) 3-way best edge Sørensen Network (b) 3-way best edge Czekanowski Network (c) Gene family enrichment network.
Fig 8.
Clustering within Brucella genus.
Subnetworks containing Brucella species constructed by selecting Brucella species and all neighbouring species nodes from (a) 3-way best edge Sørensen Network (b) 3-way best edge Czekanowski Network (c) Gene family enrichment network.
Fig 9.
Separation of Rhodobacter species.
Subnetworks containing Rhodobacter species constructed by selecting Rhodobacter species and all neighbouring species nodes from (a) 3-way best edge Sørensen Network (b) 3-way best edge Czekanowski Network (c) Gene family enrichment network.
Fig 10.
Rhodobacter and Brucella species.
Subnetworks containing Brucella and Rhodobacter species constructed by selecting Brucella and Rhodobacter species and all neighbouring species nodes from (a) 3-way best edge Czekanowski Network (b) Gene family enrichment network.