For pure samples, consensus sequence is probably most helpful (along with base/insertion counts for confidence), but for a mixed sample (e.g. wastewater), even though the base/insertion counts are probably good enough, it might still be good to output a VCF file (e.g. to make it easy to see base/insertion frequencies just at high-coverage sites that differ from the reference genome). This should be pretty easy to do: basically do the same thing as outputting the base/insertion counts, but just (1) check that the total count is above the minimum coverage, (2) compute proportions from the counts, and (3) write it in the VCF format specifically
For pure samples, consensus sequence is probably most helpful (along with base/insertion counts for confidence), but for a mixed sample (e.g. wastewater), even though the base/insertion counts are probably good enough, it might still be good to output a VCF file (e.g. to make it easy to see base/insertion frequencies just at high-coverage sites that differ from the reference genome). This should be pretty easy to do: basically do the same thing as outputting the base/insertion counts, but just (1) check that the total count is above the minimum coverage, (2) compute proportions from the counts, and (3) write it in the VCF format specifically