immuneML & Galaxy
All of immuneMLs functionalities are also available through a Galaxy web interface as a collection of Galaxy tools. We provide a YAML-based Galaxy tool that is equivalent to the CLI (command-line interface), as well as repertoire and receptor-level classification tools with an intuitive graphical user interface aimed at immunology experts without a machine learning background.
If you are unfamiliar with Galaxy, you may want to start here:
To get started using immuneML in Galaxy, you will need to add your dataset to Galaxy, which is explained in this tutorial:
Remote datasets may be fetched from VDJdb or the iReceptor Plus Gateway, see:
If you do not want to use experimental data and just want to try something out quickly, you can simulate an immune dataset:
Synthetic immune signals (representing antigen binding or disease) can be implanted in an existing dataset:
Once an immuneML dataset has been created in Galaxy, immunology experts without machine learning background can follow these instructions:
How to train immune repertoire classifiers using the simplified Galaxy interface
How to train immune receptor classifiers using the simplified Galaxy interface
Alternatively, CLI equivalent tools based on the YAML specification can be run using the following instructions:
- How to train immune repertoire classifiers using the simplified Galaxy interface
- How to train immune receptor classifiers using the simplified Galaxy interface
- How to train ML models in Galaxy
- How to apply previously trained ML models to a new AIRR dataset in Galaxy
- How to run any AIRR ML analysis in Galaxy