Python3 wrapper for running MUMmer and parsing the output.

• Introduction
• Installation
  • Required dependencies
  • Homebrew/LinuxBrew
  • Using pip
  • Running the tests
• Usage (for developers)
  • pymummer nucmer class
  • pymummer coords_file class
  • pymummer alignment class
• License
• Feedback/Issues

Runs MUMmer and parses the output.

pymummer has the following dependencies:

• MUMmer

There are a number of ways to install pymummer and details are provided below. If you encounter an issue when installing pymummer please contact your local system administrator. If you encounter a bug please log it here or email us at [email protected].

brew tap homebrew/python
brew install pymummer

pip3 install pymummer

The test can be run from the top level directory:

python setup.py test

Example showing how pymummer can be used to run nucmer on a fasta file and parse the output file to produce a set of alignment objects:

from pymummer import coords_file, alignment, nucmer
...
runner = nucmer.Runner(reference_file, query_file, results_file) 
runner.run()
file_reader = coords_file.reader(results_file)
alignments = [coord for coord in file_reader if not coord.is_self_hit()] #Remove self hits
...

Wraps the nucmer, delta-filter, show-coords and show-snps commands.

Arguments:

__ref__			reference file  
__query__			query file  
__outfile__		output file  
__min\_id__		min\_id for delta-filter command (default None)  
__min\_length__	min\_length for delta-filter command (default None)  
__breaklen__		breaklen for nucmer command (nucmer's default is 200)   
__coords\_header__	print header in show-coords output (default True)  
__maxmatch__		maxmatch for nucmer (default False)  
__show\_snps__		run show-snps (default False)  
__snps\_header__ 	print header in show-snps output (default True)  

Parses the nucmer output and populate an alignment object for each hit in the output

Check attributes of a hit, swap the reference and query, check if it's a self hit and so on

pymummer is free software, licensed under GPLv3.

Please report any issues to the issues page or email [email protected].