This component executes mash dist to find plasmids within high throughoput sequencing data, using as inputs fasta files (e.g. contigs). Then, the resulting file can be imported into pATLAS. This component calculates pairwise distances between sequences (one from the database and the query sequence). However, this process can be user for other purposes, by providing a different database than the default that is intended for plasmid searches.


  • pATLAs documentation can be found here.
  • MASH documentation can be found here.

Input/Output type

  • Input type: fasta
  • Output type: json


  • mash_distance: Sets the maximum distance between two sequences to be included in the output. Default: 0.1.


The subtraction of 1 - mash_distance can be used as an approximation to Average Nucleotide Identity (ANI). For instance a mash distance of 0.1 well correlates with ANI at 0.9 (90%).

  • pValue: P-value cutoff for the distance estimation between two sequences to be included in the output. Default: 0.05.
  • shared_hashes: Sets a minimum percentage of hashes shared between two sequences in order to include its result in the output. Default: 0.8.
  • refFile: Specifies the reference file to be provided to mash. It can either be a fasta or a .msh reference sketch generated by mash. Default: ‘/ngstools/data/patlas.msh’. If the component mash_sketch_fasta is executed before this component, this parameter will be ignored and instead the secondary link between the two processes will be used to feed this component with the reference sketch.

Published results

  • results/mashdist/: A JSON file that can be imported to pATLAS with the results from mash dist.

Published reports


Default directives

  • runMashDist:
    • container: flowcraft/mash-patlas
    • version: 1.6.0-1
  • mashDistOutputJson:
    • container: flowcraft/mash-patlas
    • version: 1.6.0-1