ERGA-consortium / pipelines

MIT License
13 stars 7 forks source link

Decontamination pipeline speedup #22

Closed tbrown91 closed 11 months ago

tbrown91 commented 1 year ago

Potentially change mapper from blast to diamond. Blast step in blobtools is prohibitively slow for large genomes. An alternative would be to use a reduced database. Need to search for an "nt-lite" like the uniprot 50/90/100 libraries, if they exist

tbrown91 commented 11 months ago

Switched to fcs-gx. I will also remove the blobtools element until the developers have incorporated the output of fcs-gx into their workflow