Click for Instructions
previous | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | Figures | next

Using additional data to optimise patterns of descent.


In some cases individual STs could be linked to (descended from) more than one ST and the eBURST diagram provides only one hypothesis for the patterns of descent. eBURST uses a set of rules that attempts to maximise the number of SLVs associated with the primary founder and the subgroup founders and there are many equally plausible minor variants of the arrangement of STs in a complex eBURST diagram.

This is somewhat similar to the construction of a phylogenetic tree, where there may be a large number of approximately equally plausible trees, which differ slightly in the branching order, but only one of them is displayed. Additional information can be used in some cases to further explore the patterns of descent implied by an eBURST diagram. For example, one SLV may have gained resistance to antibiotics, and may have diversified to become a subgroup founder, with all of its descendent STs also being antibiotic-resistant. However, there may be another ST that is resistant to the antibiotic but has not been linked by eBURST to the subgroup that includes the antibiotic-resistant STs. T
he ability to highlight all the SLVs (and DLVs) of this ST (see here) will show how it relates to the members of the antibiotic-resistant subgroup, and whether it appears to be an independently derived antibiotic-resistant ST, or is an SLV of one of the resistant subgroup and could more plausibly be linked within the subgroup of resistant STs. Similarly, other data (e.g. differences in serotypes within a clonal complex) may suggest equally parsimonious re-arrangements of the linkages between some STs shown in the eBURST diagram that make the patterns of descent more biologically plausible. The ability to show all SLVs (pink) and DLVs (blue) of any selected ST helps to explore alternative equally plausible patterns of descent (Figure 4).

Figure 4

 

eBURSTv3 has been developed and is hosted at Imperial College London