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
Sometimes the taxonomic classification is based on only a very small (<200nt), but unique, hit against a certain taxa. Then scaffold then gets assigned the taxa of that very small hit. Nevertheless, when you then do a protein blast of that same scaffold you get a much better (longer) hit against another taxa, clearly proving that is a wrong annotation.
Since it is not feasible to do a tBLASTX for everything, a solution would be to add a metric that denotes the percentage of the scaffold that is covered by the alignment and to include the BLAST bitscore. That way, people can filter for this. We might even flag these classifications as untrustworthy.
This feature is not included in the launch paper, hence it is scheduled for v1.1 (the first release post publication).
The text was updated successfully, but these errors were encountered:
Sometimes the taxonomic classification is based on only a very small (<200nt), but unique, hit against a certain taxa. Then scaffold then gets assigned the taxa of that very small hit. Nevertheless, when you then do a protein blast of that same scaffold you get a much better (longer) hit against another taxa, clearly proving that is a wrong annotation.
Since it is not feasible to do a tBLASTX for everything, a solution would be to add a metric that denotes the percentage of the scaffold that is covered by the alignment and to include the BLAST bitscore. That way, people can filter for this. We might even flag these classifications as untrustworthy.
This feature is not included in the launch paper, hence it is scheduled for v1.1 (the first release post publication).
The text was updated successfully, but these errors were encountered: