AMR detection tools

Each database we covered in step 2.3 has a specific tool to access that database and return predicted AMR patterns.
AMRFinderPlus: https://www.nature.com/articles/s41598-021-91456-0
CARD/RGI: https://pubmed.ncbi.nlm.nih.gov/36263822/
ResFinder: https://academic.oup.com/jac/article/75/12/3491/5890997
Each of the tools use some variation of BLAST plus KMC (ResFinder and CARD), or HMM (in the case of NCBI AMRfinderPlus) to identify relevant genes in a query sequence based on the genes contained in the databases. CARD and ResFinder match genes at the nucleotide level, whilst AMRFinderPlus matches genes at the protein level. Depending on the use case you may prefer to use one method over another. For instance, if you are assessing simple gene presence or absence for transmission or epidemiological studies, analysing nucleotide sequences may be sufficient. However, if functionality is also a key consideration, evaluating protein-level data might be more appropriate.
Other tools
There are a host of other tools available for AMR detection, some leveraging the existing tools outlined above, others doing completely separate things (e.g. read mapping approaches). However we would like to highlight two in particular:
- ABRicate
This tool can work on any of the databases we have covered,
allowing for comparisons between, and other databases such as
virulence factors. It does not detection mutations, only gene
presence/absence.
This tool can be accessed via Galaxy or via command line. - Kleborate
This tool is specifically for Klebsiella pneumoniae species complex.
It has its own inbuilt AMR, virulence gene, and capsule gene database.
It automatically filters AMR hits based on curated knowledge about
resistance in Klebsiella. It can be accessed via command line, or via
web interface through PathogenWatch (described below) - SCCmec
This tool looks for the SCCmec resistance genes in
Staphylococcus aureus, which cause resistance to methicillin.
The web-interface supports both reads and assemblies as input. - LREfinder
This tool detects linezolid resistance in Enterococcus, by looking
for mutations in a select set of genes. It uses a read-mapping approach,
and can be accessed via a web interface. - TB-Profiler
This is one of the main tools for AMR detection in M. tuberculosis.
It is a read mapping-based approach.
This tool can be accessed via the web interface
or via the command line. - PathogenWatch
This is a web interface where users can upload either assemblies
or reads, and the interface will detect the species of the input data.
After determining species, PathogenWatch will run a suite of tools
to characterise those genomes, depending on the input species.
For some species it provides detailed AMR results, including
phenotype prediction (see their documentation for further details).
Accessing the tools
There are various ways to interface with bioinformatics tools, but these can be categorized into three main methods, as outlined below. Each of the tools mentioned can be accessed by one or more of these methods.
Web interface
A web interface is a dedicated website designed to allow users to upload data and receive results. While such interfaces are rare for large data analyses like AMR, they do exist for certain tools.
- The CARD-RGI tool can be accessed via their website
- ResFinder also can be access via their website
Galaxy
Galaxy is an open access platform which allows researchers to undertake bioinformatics analysis using a range of tools without needing advanced computational skills. Main versions of galaxy exist, each hosted in different parts of the world and each having slightly different tools available. One core page is the https://usegalaxy.eu/ which has most tools for genomics and AMR detection.
We will provide some links and advice for using Galaxy in week 3.
- AMRFinderPlus is accessible directly in Galaxy
- CARD and ResFinder can be access this way via ABRicate, which is available on Galaxy
- The TB-Profiler and hARMonize tools (covered in a later section) are also available on Galaxy
Command line
Big data analysis requires big computing power and the ability to automate steps. The best way to do this is via the command line on your own machine or a remote server. Command line is an interface where all commands are written out; there is no point and click interface. It can be a little difficult to learn but opens up your analysis capabilities immensely.
We will provide some links and advice for learning command line interface in week 3.
All tools mentioned here have a command line interface available.
Just search for the tool name and ‘github’ to find these.
Antimicrobial Databases and Genotype Prediction: Data Sharing and Analysis
Antimicrobial Databases and Genotype Prediction: Data Sharing and Analysis
Register to receive updates
-
Create an account to receive our newsletter, course recommendations and promotions.
Register for free







