# Protocol for assessing distances in pathway space for classifier feature sets from machine learning methods

## TL;DR

This paper introduces a protocol to assess biological relationships between gene sets identified by different machine learning methods using pathway space analysis.

## Contribution

A novel protocol is introduced to evaluate the biological relevance of distinct gene sets in pathway space using the PathwaySpace R package.

## Key findings

- The protocol enables testing if seemingly different gene sets are biologically related in pathway space.
- Steps are provided for building a pathway space and calculating pathway distances between gene sets.
- Density plots and pathway distance metrics help visualize and quantify relationships between gene sets.

## Abstract

As genes tend to be co-regulated as gene modules, feature selection in machine learning (ML) on gene expression data can be challenged by the complexity of gene regulation. Here, we present a protocol for reconciling differences in classifier features identified using different ML approaches. We describe steps for loading the PathwaySpace R package, preparing input for analysis, and creating density plots of gene sets. We then detail procedures for testing whether apparently distinct feature sets are related in pathway space.

For complete details on the use and execution of this protocol, please refer to Ellrott et al.1

•Protocol to test whether distinct gene sets reflect related biology•Steps for building a pathway space for graph- or network-based distance analysis•Instructions for calculating a pathway distance metric for any pair of gene sets•Guidance on exploring relationships between gene sets in pathway space

Protocol to test whether distinct gene sets reflect related biology

Steps for building a pathway space for graph- or network-based distance analysis

Instructions for calculating a pathway distance metric for any pair of gene sets

Guidance on exploring relationships between gene sets in pathway space

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

As genes tend to be co-regulated as gene modules, feature selection in machine learning (ML) on gene expression data can be challenged by the complexity of gene regulation. Here, we present a protocol for reconciling differences in classifier features identified using different ML approaches. We describe steps for loading the PathwaySpace R package, preparing input for analysis, and creating density plots of gene sets. We then detail procedures for testing whether apparently distinct feature sets are related in pathway space.