Single sample pathway analysis in metabolomics: performance evaluation and application

Abstract

Single sample pathway analysis (ssPA) transforms molecular level omics data to the pathway level, enabling the discovery of patient-specific pathway signatures. Compared to conventional pathway analysis, ssPA overcomes the limitations by enabling multi-group comparisons, alongside facilitating numerous downstream analyses such as pathway-based machine learning. While in transcriptomics ssPA is a widely used technique, there is little literature evaluating its suitability for metabolomics. Here we provide a thorough benchmark of established ssPA methods (ssGSEA, GSVA, SVD (PLAGE), and z-score) using semi-synthetic metabolomics data, alongside the evaluation of two novel methods we propose: ssClustPA and kPCA. While GSEA-based and z-score methods outperformed the others in terms of recall, clustering/dimensionality reduction-based methods provided higher precision at moderate-to-high effect sizes. A case study applying ssPA to inflammatory bowel disease demonstrates how these methods yield a much richer depth of interpretation than conventional approaches, for example by clustering pathway scores to visualise a pathway-based patient subtype-specific correlation network. We also developed the sspa python package (freely available at https://pypi.org/project/sspa/), providing implementations of all the methods benchmarked in this study. This work underscores the value ssPA methods can add to metabolomic studies and provides a useful reference for those wishing to apply ssPA methods to metabolomics data. Author summaryPathway analysis is a computational method used to draw insights from omics data by identifying groups of molecules (biological pathways) which are important in a study. Single-sample pathway analysis is based on the same principles as conventional pathway analysis but allows researchers to compute the enrichment of pathways at an individual-sample level. This enables pathway-based patient stratification and facilitates a multitude of downstream analyses based on pathways, such as machine learning, or pathway-based visualization which would not be possible using conventional approaches. In this work we investigated the application of single-sample pathway analysis to metabolomics data, a field in which it has not been widely used to date. We first evaluated the most popular methods for single-sample pathway analysis using simulated metabolomics data, as well as two novel methods we developed. Following these tests, we used metabolomics data from inflammatory bowel disease patients to demonstrate how single-sample pathway analysis methods can be used to infer novel pathway-based insights with the advantage of discriminating between disease subtypes. Overall, our analysis provides readers with information on the performance of single-sample pathway analysis methods, whilst highlighting the potential of such methods in metabolomics research.