How to apply previously trained ML models to a new AIRR dataset in Galaxy

After having trained ML models to a given dataset, these models can be applied to a new dataset using the Galaxy tool Apply machine learning models to new data. If you instead want to train new ML models, see the tutorials for training ML models for receptor and repertoire classification using the easy Galaxy interfaces, or the more versatile YAML-based tool for training ML models.

An example Galaxy history showing how to use this tool can be found here.

Creating the YAML specification

This Galaxy tool takes as input an immuneML dataset from the Galaxy history, a model training output .zip, and a YAML specification file.

The YAML specification should use the MLApplication instruction. The .zip file contains all information immuneML needs to apply the same preprocessing and encoding as to the original dataset, and to make predictions using the same ML model. More details are explained in the tutorial How to apply previously trained ML models to a new dataset.

When writing an analysis specification for Galaxy, it can be assumed that all selected files are present in the current working directory. A path to an additional file thus consists only of the filename.

A complete YAML specification for applying ML models to a new dataset is shown here:

    dataset: # user-defined dataset name
      format: ImmuneML # the default format used by the 'Create dataset' galaxy tool is ImmuneML
        path: dataset.iml_dataset # specify the dataset name, the default name used by
                                  # the 'Create dataset' galaxy tool is dataset.iml_dataset
    type: MLApplication
    dataset: dataset
    config_path: # the name of the ML model
    number_of_processes: 4

Tool output

This Galaxy tool will produce the following history elements:

  • Summary: ML model application: a HTML page that allows you to browse through all results, including predictions made on the new dataset.

  • Archive: ML model application: a .zip file containing the complete output folder as it was produced by immuneML. This folder contains the output of the MLApplication instruction such as the predictions on the new dataset. Furthermore, the folder contains the complete YAML specification file for the immuneML run, the HTML output and a log file.