New argument --best N keeps only the best N hits for each query sequence, as ranked by absolute number of mismatches from the query.
I've been comparing the before and after (with and without --best 2) for a particular data set generated by the capture pipeline (classifying individual query sequences) and the results appear pretty promising. You can compare the before and after classifications here:
New argument
--best N
keeps only the best N hits for each query sequence, as ranked by absolute number of mismatches from the query.I've been comparing the before and after (with and without
--best 2
) for a particular data set generated by the capture pipeline (classifying individual query sequences) and the results appear pretty promising. You can compare the before and after classifications here:/mnt/disk2/molmicro/working/tland9/2016-02-25_bioy_slashname_filtering/classifications-nobest /mnt/disk2/molmicro/working/tland9/2016-02-25_bioy_slashname_filtering/classifications-best2
Potential Feedback: