Improve the convergence of MCMC-based inference framework of mutation trees applying the Optimum Branching algorithm on the trees sampled from the posterior
- Clone the COBtree repository
git clone https://github.com/DavideMaspero/COBtree.git
- Install Docker following the instruction at this link: (https://www.docker.com/get-started).
- Get COBtree docker image. There are two options to do this:
- Download it from Docker Hub (https://hub.docker.com/u/dcblab)
docker pull dcblab/cobtree_img:latest - Create it from docker file
cd docker docker build -t cobtree_img .
- Download it from Docker Hub (https://hub.docker.com/u/dcblab)
- Parameters can be changed by editing the nextflow.config file in COBtree directory (plese, do not rename it)
- Run the docker image iteratively by specifing as source directory the local absolute path where COBtree repository is located. The SCITE inferences are executed in parallel. So, please replace n_cpus with the number of CPUs available.
docker run -it --rm --cpus n_cpus --mount type=bind,source=/local-dir-absolute-path/COBtree/,target=/nextflow_runs/ dcblab/cobtree_img:latest
- Inside the cobtree_img, move to the nextflow_run directory
cd nextflow_run
- Run the Simulation
nextflow run COBtree_analyses.nf
The results (plots and metric scores) are stored in the COBtree/results directory.