Abstract

In this tutorial, we will build a new immunotherapy-response classifier with Galaxy Tabular Learner using a comprehensive dataset of patients treated with immune checkpoint blockade (ICB) and non-ICB-treated patients across 18 solid tumor types. The goal is to accurately predict patient responses to the treatment.

About This Material

This is a Hands-on Tutorial from the GTN which is usable either for individual self-study, or as a teaching material in a classroom.

Questions this will address

• How can the Tabular Learner in Galaxy be used to reconstruct the LORIS LLR6 logistic regression model described by Chang et al. (2024) using the same dataset and predictor set?
• How should the decision-threshold parameter be configured in Tabular Learner (default versus optimized, and the selection criterion) to align predictions with the intended clinical operating point?
• Which components of the Tabular Learner report best support a transparent comparison between the Galaxy-built model and the performance reported in the LORIS paper?

Learning Objectives

• Use a large dataset of immune checkpoint blockade (ICB)-treated and non-ICB-treated patients across 18 solid tumor types, encompassing a wide range of clinical, pathologic and genomic features to build a Machine Learning Model.
• Build a Machine Learning model using Tabular Learner in Galaxy.
• Evaluate the model for robustness by comparing it with the original LORIS LLR6 model published by Chang et al., 2024.