Closed csmiller closed 11 months ago
@csmiller Currently we only use wildcard or card variants with the rgi bwt
program. Yes, the variants do improve predictions of AMR genes in metagenomes, at least that was our conclusion in this poster https://github.com/arpcard/asm_microbe_2019_rgi/blob/master/ASM_MICROBE_2019_POSTER_A_R_RAPHENYA.pdf
First, thanks for a great tool and a great resource.
We are annotating already inferred protein sequences from metagenome-assembled genomes. I notice that, at least in 2021 (issue #156), rgi main run in protein mode could only search the canonical CARD database, and not the Resistomes & Variants (WildCARD) data.
With the most recent versions of RGI and CARD/WildCARD Is it possible and/or advisable to also search protein sequences against the WildCARD database? I am wondering if this would increase sensitivity in a metagenomic protein dataset in particular, without increasing false positives too much.
If possible, what would the appropriate rgi load command look like before running rgi main -t protein ... ?
If not advisable and/or possible, if not too much trouble could you briefly elaborate on why, please?
Thanks!