UnPaSt: unsupervised patient stratification by biclustering of omics data
Abstract
Unsupervised patient stratification is essential for disease subtype discovery, yet, despite growing evidence of molecular heterogeneity of non-oncological diseases, popular methods are benchmarked primarily using cancers with mutually exclusive molecular subtypes well-differentiated by numerous biomarkers. Evaluating 22 unsupervised methods, including clustering and biclustering, using simulated and real transcriptomics data revealed their inefficiency in scenarios with non-mutually exclusive subtypes or subtypes discriminated only by few biomarkers. To address these limitations and advance precision medicine, we developed UnPaSt, a novel biclustering algorithm for unsupervised patient stratification based on differentially expressed biclusters. UnPaSt outperformed widely used patient stratification approaches in the de novo identification of known subtypes of breast cancer and asthma. In addition, it detected many biologically insightful patterns across bulk transcriptomics, proteomics, single-cell, spatial transcriptomics, and multi-omics datasets, enabling a more nuanced and interpretable view of high-throughput data heterogeneity than traditionally used methods.
Keywords
Cite
@article{arxiv.2408.00200,
title = {UnPaSt: unsupervised patient stratification by biclustering of omics data},
author = {Michael Hartung and Andreas Maier and Yuliya Burankova and Fernando Delgado-Chaves and Olga I. Isaeva and Alexey Savchik and Fábio Malta de Sá Patroni and Jens J. G. Lohmann and Daniel He and Casey Shannon and Jan-Ole Schulze and Katharina Kaufmann and Zoe Chervontseva and Farzaneh Firoozbakht and Anne Hartebrodt and Niklas Probul and Olga Tsoy and Alexandra Abisheva and Evgenia Zotova and Kavya Singh and Kristel Van Steen and Malte Kuehl and Victor G. Puelles and David B. Blumenthal and Martin Ester and Tanja Laske and Jan Baumbach and Olga Zolotareva},
journal= {arXiv preprint arXiv:2408.00200},
year = {2026}
}
Comments
Substantially revised version with additional analyses