ConsensusCore is a C++ library containing several consensus- and variant- calling algorithms used in circular consensus sequencing (CCS) analysis and in genomic consensus/variant analyses. Several of the algorithms are based on generic multiple sequence alignment methods, while the Quiver method is based on a PacBio-specific error model. ConsensusCore exposes language bindings to C# and Python, enabling integration into the Pipeline and secondary analysis workflows. Note that ConsensusCore is a library, and does not contain any driver programs. Most users will be interested in installing GenomicConsensus, which contains the variantCaller.py program. See installation instructions.

To install in your virtualenv (or your global python environment, if you wish), run

$ python setup.py install --swig=/path/to/swig_executable --boost=/boost/root

If you want to play with just the C++ library or the C# bindings, you can build like so:

$ make              # build C++ library ...
$ make tests        # ... and test it.
$ make csharp       # build CSharp bindings
$ make csharp-demo  # run a C# demo

The best place to get started is the BrownBag2012 presentation under the doc/ folder. Then, to see how to write code using ConsensusCore, take a look at quiver/demo.py under the (separate) GenomicConsensus project.