Hi NanoSeq devs,

Thank you for making your software available for all to use! We encountered an issue where in rare cases/loci (e.g. chrM:0-1) a read bundle would have an empty trinucleotide column (field 4 in the dsa .bed outputs), which would then be interpreted as a truncated .bed as two tab characters don't get interpreted as two column separators in awk without an explicit FS variable being set. The offending line is:

As far as we can tell in our .bams we don't believe we handled the files inappropriately, the only thing we can think of is that nearly all the mappings at this locus are soft-clipped, so if the column is being populated by the mapped sequence at this position then it would explain why the field is empty.