AuriClass is a small tool which predicts Candida auris clade based on Mash distances from reference genomes. It accepts fastq or fasta files. Analysis typically takes a minute for Fastq data and a couple of seconds for Fasta data.
Documentation is available from https://rivm-bioinformatics.github.io/auriclass.
The easiest way to install AuriClass is through mamba/conda:
mamba create -n env_auriclass -c bioconda -c conda-forge auriclass
conda activate env_auriclass
Alternatively, biocontainers has built a container from the AuriClass bioconda package. This container can be used with different container software, e.g. Singularity:
singularity pull docker://quay.io/biocontainers/auriclass:0.5.1--pyhdfd78af_0
Please note that auriclass:latest is not defined, so make sure to specify the latest bioconda version and build.
Running AuriClass on only the forward reads gives the best results, as the reverse reads are usually more noisy:
auriclass Candida_auris_R1.fq.gz
But it can also be run on an arbitrary number of fastq files of the same organism:
auriclass Candida_auris_R1.fq.gz Candida_auris_R2.fq.gz Candida_auris_unpaired.fq.gz
or with a fasta file:
auriclass Candida_auris.fasta.gz
A standard analysis creates two files:
report.tsvwhich contains the clade prediction, closest reference sample and QC checks. The report contains the default "isolate" as sample name.report.YYYY-mm-dd_HH-MM-SS.logwhich contains messages written to STDERR with additional data.