Hello,
In my lab, we do trnL and 12S sequencing, using a high-fidelity polymerase for trnL and a low-fidelity polymerase for 12S. For trnL, single base pair differences may result in vast phenotypic diversity, while for 12S, a cow might have 20 distinct ASVs. This makes assignSpecies() perfect for trnL (exact matching) and assignTaxonomy() perfect for 12S (Bayesian classifier).
However, we use references with taxonomy headers, and we're wondering if it'd be fine to create a function to merge assignTaxonomy()'s taxonomy header compatibility with assignSpecies()'s assignment algorithm, since currently, using accession headers and a taxonomizr-formatted SQL file takes a lot longer and seems unnecessarily roundabout when we've already made a trnL reference with taxonomy headers. We were also curious about why the two functions are built so differently, the potential advantages of accession headers given these drawbacks, and any potential considerations for our Frankenstein function. Thanks!
Hello,
In my lab, we do trnL and 12S sequencing, using a high-fidelity polymerase for trnL and a low-fidelity polymerase for 12S. For trnL, single base pair differences may result in vast phenotypic diversity, while for 12S, a cow might have 20 distinct ASVs. This makes assignSpecies() perfect for trnL (exact matching) and assignTaxonomy() perfect for 12S (Bayesian classifier).
However, we use references with taxonomy headers, and we're wondering if it'd be fine to create a function to merge assignTaxonomy()'s taxonomy header compatibility with assignSpecies()'s assignment algorithm, since currently, using accession headers and a taxonomizr-formatted SQL file takes a lot longer and seems unnecessarily roundabout when we've already made a trnL reference with taxonomy headers. We were also curious about why the two functions are built so differently, the potential advantages of accession headers given these drawbacks, and any potential considerations for our Frankenstein function. Thanks!