Not an issue but a question re reference database

[hydrahamster]

Problem Description

Hi there, I would love to use smhlast. However we are doing de novo transcriptome annotation, historically against the nr ncbi database. We don't have the luxury of closely related sequenced organisms, so this seems the best database to use. But using BLAST on the cluster is incredibly slow.
Does the nr database present the same problem as uniref90, the mixed protein database you mentioned as an example?

Thanks!