Results of assemblies and mappings of five Hong Kong swine vesicular disease virus (SVDV) isolates.

First row shows the names of the five isolates, passage history, and harvest date. Second row shows the number of reads assigned to each isolate after they were separated using MID barcodes, and the number of these that had unique sequences (i.e. with duplicates removed), which were then kept for assembly/mapping. A number of reads is also lost during de-multiplexing, however this is even smaller than the total number of duplicates (not shown). Row three shows the mean, the standard deviation and the maximum of the read lengths of the unique reads. The fourth row shows results from de novo assembly, including number of reads mapped and minimum, maximum, mean, and standard deviation figures for the depth of coverage of the contig produced. The following five rows show results from iterative mapping against the five selected starting reference sequences (two SVDV sequences, a coxsackievirus B5, a coxsackievirus B3, and a coxsackievirus B6 sequence). The three isolates with successful de novo assemblies were iteratively mapped against one or two of these sequences for validation (see Figure 1), whereas the remaining two isolates were iteratively mapped against all five starting references. For each of these two isolates the final mapping statistics can be seen to be virtually identical across all five starting references, clearly indicating that convergence of the iterative mapping process has been obtained. The final row shows the total number of ‘polymorphic’ sites obtained when the consensus sequence from assembly (if successful) and all of the mappings are multiple-aligned for each isolate (in parenthesis is shown whether any of the ‘polymorphisms’ are due to base calls where a non-redundant call in one sequence matches the category of a redundant call in the other).