search

bokulich-lab / nf-ducken

Workflow to process amplicon meta-analysis data, from NCBI accession IDs to taxonomic diversity metrics.
3 stars 2 forks source link

Manage optimal sample-primer classification out of Cutadapt #83

Closed lina-kim closed 5 months ago

lina-kim commented 9 months ago

And, as a result, reintegrate the COMBINE processes that had to be removed due to errors.

lina-kim commented 6 months ago

Tabled as #97 removed the need for optimization and selection.

lina-kim commented 6 months ago

Reopening. While native Cutadapt chooses an optimal primer from given inputs, it does not distinguish between "best primer" and "primer has already been trimmed". The latter would have lower alignment percentages than expected (far lower than the 99% given by regular untrimmed primers on FASTQs).

Time to break out the parser again? We may need to remove the default --discard-untrimmed reads option to incorporate at a later step.