Install immuneML with a package manager
Create a directory for immuneML and navigate to the directory:
Create a virtual environment using conda. immuneML has been tested extensively with Python versions 3.7 and 3.8, but not 3.9. immuneML depends on PyTorch, which fails to install with Python 3.9. To create a conda virtual environment with Python version 3.8, use:
conda create --prefix immuneml_env/ python=3.8 setuptools=50.3.2
setuptools=50.3.2 is a temporary workaround for the issue described here: When running immuneML, I get “ModuleNotFoundError: No module named ‘init’”.
Activate the created environment:
conda activate immuneml_env/
Install basic immuneML including Python dependencies using pip:
pip install immuneML
pip install immuneML[TCRdist]
See also this question under ‘Troubleshooting’: I get an error when installing PyTorch (could not find a version that satisfies the requirement torch)
5. Optionally, if you want to use the DeepRC ML method and and corresponding DeepRCMotifDiscovery report, you also
have to install DeepRC dependencies using the
Important note: DeepRC uses PyTorch functionalities that depend on GPU. Therefore, DeepRC does not work on a CPU.
To install the DeepRC dependencies, run:
pip install -r requirements_DeepRC.txt --no-dependencies
6. Optionally, if you want to use the CompAIRRDistance or CompAIRRSequenceAbundance encoder, you have to install the C++ tool CompAIRR.
The easiest way to do this is by cloning CompAIRR from GitHub and installing it using
make in the main folder:
git clone https://github.com/uio-bmi/compairr.git
If such installation is unsuccessful (for example if you do not have the rights to install CompAIRR via make), it is also possible to directly provide the path to a CompAIRR executable as a parameter to CompAIRRDistance or CompAIRRSequenceAbundance encoder.
To validate the installation, run:
The output should look like this:
usage: immune-ml [-h] [--tool TOOL] specification_path result_path
immuneML command line tool
specification_path Path to specification YAML file. Always used to define
result_path Output directory path.
-h, --help show this help message and exit
--tool TOOL Name of the tool which calls immuneML. This name will be
used to invoke appropriate API call, which will then do
additional work in tool-dependent way before running
To quickly test out whether immuneML is able to run, try running the quickstart command:
This will generate a synthetic dataset and run a simple machine machine learning analysis on the generated data.
The results folder will contain two sub-folders: one for the generated dataset (
synthetic_dataset) and one for the results of the machine
learning analysis (
machine_learning_analysis). The files named specs.yaml are the input files for immuneML that describe how to generate the dataset
and how to do the machine learning analysis. The index.html files can be used to navigate through all the results that were produced.