You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We are planning to benchmark a set of metabarcoding tools and I thought to use OPAL. When investigating the tool, I thought one could use it for metabarcoding assessment just as well as long as one can provide a read-taxonomy gold-standard it should be able to compute metrics for i.e. unifrac / recall the same way. Any insight from you why not just using it for metabarcoding as well ? Sure the OTU clustering can not be accessed, but anything regarding the taxonomy should just work the same way... ?
The text was updated successfully, but these errors were encountered:
If you need to assess taxonomic assignment per read, you could use AMBER (https://github.com/CAMI-challenge/AMBER). Would that work? If not, could you describe what types of files you need to assess, like formats and contents? I am not so familiar with metabarcoding assessment.
Dear @fernandomeyer thank you very much for the quick respond, yes I thought about AMBER as well and we will use it for binning evaluation, however for amplicon analysis the main result of the analysis is usually the overall community composition, therefore we are looking in
to OPAL. I thought the actual outputs of amplicon / shotgun community taxonomy are similar in the sense, that both have read-per-taxa-rank, so I think OPAL should work. I am attaching a standard output of amplicon analysis. OTU_example.txt
We are planning to benchmark a set of metabarcoding tools and I thought to use OPAL. When investigating the tool, I thought one could use it for metabarcoding assessment just as well as long as one can provide a read-taxonomy gold-standard it should be able to compute metrics for i.e. unifrac / recall the same way. Any insight from you why not just using it for metabarcoding as well ? Sure the OTU clustering can not be accessed, but anything regarding the taxonomy should just work the same way... ?
The text was updated successfully, but these errors were encountered: