Spack

Having a consistent installation of software is important to use tumorcode. Spack is a flexible package manager supporting every hardware architecture. Initially this part was started to run tumorcode on Mac OS.

Getting spack

The github page of spack is can be found here: https://github.com/LLNL/spack Please read the documentation carefully, since there are several pitfalls depending on your hardware and operating system.

Adding tumorcode

We denote the spack installation path by SPACK_ROOT. After installation supported packages can be found in the folder $SPACK_ROOT/var/spack/repos/builtin/packages/. There you have to create a new folder named tumorcode and copy the file package.py distributed by tumorcode to this folder. If you are familiar with spack, that was it. For newbies should continue reading.

Environment manager

Computing cluster often provide software in different versions. Sometimes these version are dependent on your environment variable. For convenience the variable could be manged by software. We use the dotkit manager documented at https://computing.llnl.gov/?set=jobs&page=dotkit For usage of tumorcode we had to do the following tasks

• enable shell integration
• tell the console to use spack's python by spack use python
• activate the follwoing packages by spack activate "package name" ( see https://spack.readthedocs.io/en/latest/basic_usage.html#extensions-python-support
• py-numpy, py-h5py,

Usage

• Make sure you use spack's python
• Use tumorcode scripts from spack installation path