Accelerating the rRNA filtering step
sortmerna appears to be a relatively slow program for the metatranscriptomic files we are typically processing.
An acceleration of this step would be nice!
There do not seem to be too many alternatives out there though: https://omictools.com/rrna-filtering-category Some of them are web-based, hence do not really represent an alternative.
rrnafilter would be one, but from superficially scanning over it, it seems to be based on k-mer abundances.
While this enables the identification of candidate rRNA reads without the need for reference sequences, it could also lead to false positives, i.e., non-rRNA gene-derived sequences of high (or divergent) abundance) that are wrongly classified as rRNA gene-derived.
Not sure how the tool was tested in the original publication. A dedicated test might be required, though.