AMRrules
Interpretive standards for AMR genotypes
Organism-specific interpretation of antimicrobial susceptibility testing (AST) data is standard in clinical microbiology, with rules regularly reviewed by expert committees of EUCAST and CLSI. We aim to provide an analagous resource to support organism-specific interpretation of antimicrobial resistance (AMR) genotypes derived from pathogen whole genome sequence (WGS) data.
AMRrules encode organism-specific rules for the interpretation of AMR genotype data, and are curated by organism experts belonging to ESGEM-AMR, a working group of ESGEM, the ESCMID Study Group on Epidemiological Markers. The rule specification is available in this Google sheet (v0.5, guidance on tab 2).
This AMRrules Python package includes the rules themselves (see rules/ directory) as well as code to apply the rules to interpret AMR genotypes (currently limited to AMRfinderplus output), generating informative genome reports that capture expert knowledge about how core and acquired genes and mutations contribute to antimicrobial susceptibility.
We are focusing early development on compatibility with NCBI resources (i.e. the AMRfinderplus genotyping tool, and the associated NCBI databases including AMR refgene, AMR Reference Gene Hierarchy, and the Reference HMM Catalog). In future we plan for interoperability with CARD and ResFinder (and other tools based on these), using hAMRonization.
Initial rule curation has focused on defining rules for the interpretation of core genes and expected resistances, but acquired genes and mutations are included for some organisms already and will be added to others as the necessary data to define them accurately is accumulated and curated by the ESGEM-AMR working group.
Organism-specific rules
Full specification: AMRrules spec v0.5. Note this includes several additional fields beyond those pictured above, including NCBI and CARD ARO accessions to uniquely identify genes; details of the breakpoints and standards used; evidence codes, grades and limitations; and a rule annotation note.
Available rules
Rule curation is a work in progress, under active development by the ESGEM-AMR Working Group.
Currently available rule sets are in the rules/ directory of this repository, named by organism. In this beta release they focus mainly on core genes and expected resistances, however acquired genes and mutations are included for some organisms already and will be added to others as the necessary data to define them accurately is accumulated and curated by the ESGEM-AMR working group.
- Acinetobacter baumannii
- Enterobacter
- Enterococcus faecalis
- Enterococcus faecium
- Escherichia coli
- Klebsiella pneumoniae
- Neisseria gonorrhoeae (acquired resistances, based on analysis of geno-pheno data)
- Pseudomonas aeruginosa
- Salmonella
- Staphylococcus aureus
- Yersinia
Interpreting AMRfinderplus genotype results with AMRrules
Package installation
The only dependency is Python, v3.12 or higher
conda create -n amrrules_beta -c bioconda python=3.12
conda activate amrrules_beta
conda install pip
git clone https://github.com/interpretAMR/AMRrules
cd AMRrules
make dev
Use AMRrules to interpret AMRfinderplus results for a single genome
amrrules --input tests/data/input/test_ecoli_genome.tsv --output_prefix test_ecoli_genome --organism 's__Escherichia coli'
amrrules --input tests/data/input/test_kleb_SGH10.tsv --output_prefix test_kleb_SGH10 --organism 's__Klebsiella pneumoniae'
Use AMRrules to interpret results for multiple genomes of different organisms (test data)
amrrules --input tests/data/input/test_data_amrfp_multiSpp.tsv --output_prefix test_multispp --organism_file tests/data/input/test_data_sppCalls.tsv
Example command to run AMRfinder plus (note the –print_node option) and interpret the output with AMRrules
amrfinder -n Kpn1.fasta --plus --print_node --name Kpn1 --organism Klebsiella_pneumoniae > Kpn1_AMRfp.tsv
amrrules --input Kpn1_AMRfp.tsv --output_prefix Kpn1 --organism 's__Klebsiella pneumoniae'
Example interpreted genotype report
Example interpreted summary output for wildtype Klebsiella pneumoniae str. SGH10
Note the image includes only selected fields from the AMRfinderplus report, for illustration purposes. The AMRrules output file includes all columns in the input genotype file, with the additional AMRrules annotation columns (highlighted in purple) added to the end.
Example file, generated using test commands above: test_kleb_SGH10_interpreted.tsv
Example file with multiple genomes from different species, generated using test commands above: test_multiSpp_interpreted.tsv
Example generic genome report
Work in progress, not yet available in the beta release
Contributors
The AMRrules concept was initially workshopped by members of the Holt lab at London School of Hygiene and Tropical Medicine and further developed in collaboration with Jane Hawkey at Monash University. The AMRrules specification was developed by the ESGEM-AMR Data & Tools group, and the rules curated by the ESGEM-AMR Working Group (see list of members), chaired by Natacha Couto (ESGEM Chair). Code was developed by Jane Hawkey and Kat Holt.