SiCloneFitIO is a python wrapper with cmd line tool to covert data matrix for use of siCloneFit from and to pandas DataFrame and plot imputed matrix overlay with raw data in clustermap.
SiCloneFit is a Bayesian method for joint inference of tumor clones, clonal genotypes and clonal phylogeny from noisy mutation profile of single cells. Given an imperfect noisy genotype matrix from single cells, SiCloneFit employs a Gibbs sampling algorithm to sample from the posterior distribution of the probabilistic graphical model. Each sample contains the cluster indicator for all cells, genotypes of each clone, a clonal phylogeny that represents the genealogical relationships and error rate values.
- (Create an virtual environment of choice)
- Download and install siclonefitio
git clone https://github.com/zztin/siCloneFitIO.git
cd ./siclonefit
python setup.py install
- Run tests for siclonefitio
cd tests/
* test007 test the complete functionality, takes around 5 mins
python -m unittest test_siclonefitio.TestSiclonefitio.test_007_siclonefitIO
* If failed, run all tests. This takes around 10 mins.
python -m unittest
- install other dependencies for siclonefit if needed (check https://bitbucket.org/hamimzafar/siclonefit/src/master/)
siclonefit -j ../hamimzafar-siclonefit/SiCloneFiTComplete.jar -s {input_snvmatrix_pd_dataframe_pickle} \
-cn ../test_data/cnv.pickle.gz -o {output_dir} -n test1 -mm {minMeasurement} -mp {minPresence}
# run test data. the output test_out_local/ should be identical to test_out/ on github repo.
siclonefit -j ../hamimzafar-siclonefit/SiCloneFiTComplete.jar -s ../test_data/test1.pickle \
-cn ../test_data/cnv.pickle.gz -o ../test_out_local/ -n test1 -mm 1 -mp 1
This package was created with Cookiecutter_ and the audreyr/cookiecutter-pypackage_ project template.
.. _Cookiecutter: https://github.com/audreyr/cookiecutter
.. _audreyr/cookiecutter-pypackage: https://github.com/audreyr/cookiecutter-pypackage