Taming the AMR beast
abriTAMR is an AMR gene detection pipeline that runs AMRFinderPlus on a single (or list ) of given isolates and collates the results into a table, separating genes identified into functionally relevant groups.
abriTAMR is accredited by NATA for use in reporting the presence of reportable AMR genes in Victoria Australia.
- Acquired resistance mechanims in the form of point mutations (restricted to subset of species)
- Streamlined output.
- Presence of virulence factors
abriTAMR requires AMRFinder Plus, this can be installed with conda
alone or as part of the abriTAMR conda
installation (see below).
conda create -n abritamr -c bioconda ncbi-amrfinder
conda activate abritamr
amrfinder -U (to download and install recent DB)
pip3 install abritamr
abritamr run --help
optional arguments:
-h, --help show this help message and exit
--contigs CONTIGS, -c CONTIGS
Tab-delimited file with sample ID as column 1 and path
to assemblies as column 2 OR path to a contig file
(used if only doing a single sample - should provide
value for -pfx). (default: )
--prefix PREFIX, -px PREFIX
If running on a single sample, please provide a prefix
for output directory (default: abritamr)
--species {Acinetobacter_baumannii,Campylobacter,Enterococcus_faecalis,Enterococcus_faecium,Escherichia,Klebsiella,Salmonella,Staphylococcus_aureus,Staphylococcus_pseudintermedius,Streptococcus_agalactiae,Streptococcus_pneumoniae,Streptococcus_pyogenes,Vibrio_cholerae}, -sp {Acinetobacter_baumannii,Campylobacter,Enterococcus_faecalis,Enterococcus_faecium,Escherichia,Klebsiella,Salmonella,Staphylococcus_aureus,Staphylococcus_pseudintermedius,Streptococcus_agalactiae,Streptococcus_pneumoniae,Streptococcus_pyogenes,Vibrio_cholerae}
Set if you would like to use point mutations, please
provide a valid species. (default: )
--jobs JOBS, -j JOBS Number of AMR finder jobs to run in parallel.
(default: 16)
You can also run abriTAMR in mdu
mode, this will output a spreadsheet which is based on reportable/not-reportable requirements in Victoria. You will need to supply a quality control file (comma separated) (-q
), with the following columns:
- ISOLATE
- SPECIES_EXP (the species that was expected)
- SPECIES_OBS (the species that was observed during the quality control analysis)
- TEST_QC (PASS or FAIL)
--sop
refers to the type of collation and reporting pipeline
- general
- standard reporting structure for aquired genes, output as reportable and non-reportable
- salmonella
- Inferred AST based on validation undertaken at MDU
--qc QC, -q QC Name of checked MDU QC file. (default: )
--runid RUNID, -r RUNID
MDU RunID (default: Run ID)
--matches MATCHES, -m MATCHES
Path to matches, concatentated output of abritamr
(default: summary_matches.txt)
--partials PARTIALS, -p PARTIALS
Path to partial matches, concatentated output of
abritamr (default: summary_partials.txt)
--sop {general,salmonella}
The MDU pipeline for reporting results. (default:
general)