Use cases and comparison to other tools
We tested our software on publicly available data from Maestri et al.
(2019) and Wurzbacher et al. (2019), and compared the accuracy of
respective consensus sequences generated in the two studies to those
reconstructed with NGSpeciesID. To measure accuracy, we aligned the
consensus sequences to the Sanger sequence using BLAST (Altschul et al.,
1990) and calculated accuracy as the sum of all matches in the alignment
divided by the alignment length. We chose the software solution
presented in Wurzbacher et al., (2019) for our comparison as it is
currently, to our knowledge, the only one that can handle both PacBio
and ONT sequencing reads. We further compared our result to the ONTrack
software (Maestri et al., 2019) developed for ONT data specifically. In
both comparisons we carried out polishing within NGSpeciesID using
Medaka (https://github.com/nanoporetech/Medaka) and Racon (Vaser et al.,
2017).