BURST (Based Upon Related Sequence Types) is a novel clustering algorithm designed for use on microbial MLST data developed by Ed Feil (University of Bath). The approach specifically examines the relationships within clonal complexes while the relationships between different clonal complexes are ignored.  This implementation will work with datasets of four or more loci.

BURST requires allelic profile data only, but these must contain ST numbers (select 'Include Sequence Types' on 'Load Profiles' dialog).  To run click 'Analysis ... Lineage Assignment ... BURST'.

You will initially be faced with a dialog window, offering a number of options.  The first of these is the definition of a group.  Here you enter the number of loci that members of a group must share with at least one other member of the group.   For MLST data based on 7 loci, it has been found that a cut-off point of 5 identical loci maximises the inclusion of strains belonging to a single clonal complex, while excluding those which do not.

burst dialog

The next series of options concerns the display of a summary view.  If an ancestral type is identified for a group, a summary view is displayed.  This shows a series of concentric circles surrounding the ancestral type.  The middle circle contains STs that vary at a single locus (single-locus variants - SLVs) and the outer contains STs that vary at two loci (double-locus variants - DLVs).  Outside of this circle are satellite STs that vary by more than two loci from the central type.  STs may be linked by lines showing whether they're SLVs or DLVs of other STs.  DLV links are not shown between two SLVs of the ancestral type, since these will all be DLVs (or occassionally SLVs) of each other.  The dialog window provides the option to select the colours and line styles of links, or whether or not to show SLV or DLV links.


The program outputs a series of tables showing the members of each group and the number of SLVs, DLVs and satellites each ST has.   If an ST has been identified as an ancestral type, it's ST number will have an asterisk next to it and there will be a summary view, as above.

There may also be a link to an expanded view, which is useful for large diverse groups where there may be multiple ancestral types interlinked to each other.  For these, the summary view may be overly cluttered. The link will be available if the number of required identical loci in your group definition is within two loci of the total number of loci.  This limitation is imposed because the algorithm used to draw the expanded view uses the number of SLVs and DLVs that each ST has to order the diagram.  Clearly there may be some STs with no SLVs or DLVs if the group definition is more than two loci fewer than the total number of loci.  The expanded view is produced in Postscript format and a hyperlink will point to this file.  If your system has Ghostscript installed (see system requirements), this may start up automatically and display the diagram upon clicking the hyperlink.