BioCMA

Consensus Multiple Alignment format, using Biopython alignments.

I/O support and relevant functionality for the Consensus Alignment Format (CMA). This format represents protein sequence alignments. It is used by a few tools by Dr. Andrew F. Neuwald, notably CHAIN and MAPGAPS.

Biopython objects and conventions are used where possible.

Installation

This is an ordinary Python package. You can install it from source with the setup.py script:

python setup.py build
python setup.py install

Stable releases are uploaded to PyPI as well, so you can install BioFrills with Python package managers:

pip install biofrills

Or:

easy_install biofrills

Some scripts that I find useful are in the scripts/ directory. By default these are not installed, but you can include them by uncommenting the line in setup.py that starts with scripts=glob...

Alternatively, you can just copy those scripts into another directory in your $PATH.

What can the CMA format do for me?

• Like the A2M and Stockholm formats, alignments are shown with insertions as lowercase characters and deletions are dashes
• Like the A3M format, alignments are pairwise versus a profile (or "consensus" sequence), which also dictates which sites are indels. By compressing the insert columns, a large of alignment of many divergent (but related) sequences can be shown without filling it will mostly gap characters, as Stockholm can.
• Like Stockholm, but unlike A2M and A3M, more than one alignment can be contained in a single file.
• Typically, an ungapped consensus sequence will be included as the first sequence.
• A FASTA-like header contains additional, optional fields for the number of leading and trailing sites and NBCI taxonomy codes

Who uses CMA?

The CMA format appears to have been invented by Dr. Andrew Neuwald at the University of Maryland, and is used in these programs:

• MAPGAPS: http://mapgaps.igs.umaryland.edu/
• CHAIN: http://chain.igs.umaryland.edu/

Should I use CMA in my own work?

Unless you're working with MAPGAPS or CHAIN, no.