Summary of Unpast: Unsupervised Patient Stratification by Differentially Expressed Biclusters in Omics Data, By Michael Hartung et al.
UnPaSt: unsupervised patient stratification by differentially expressed biclusters in omics data
by Michael Hartung, Andreas Maier, Fernando Delgado-Chaves, Yuliya Burankova, Olga I. Isaeva, Fábio Malta de Sá Patroni, Daniel He, Casey Shannon, Katharina Kaufmann, Jens Lohmann, Alexey Savchik, Anne Hartebrodt, Zoe Chervontseva, Farzaneh Firoozbakht, Niklas Probul, Evgenia Zotova, Olga Tsoy, David B. Blumenthal, Martin Ester, Tanja Laske, Jan Baumbach, Olga Zolotareva
First submitted to arxiv on: 31 Jul 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Genomics (q-bio.GN)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This research paper proposes a novel approach to patient stratification for complex diseases, such as cancer and asthma. The existing methods have been benchmarked primarily on cancer omics data and are limited by requiring multiple biomarkers to characterize distinct subtypes. The authors contribute to the field by evaluating 22 unsupervised patient stratification methods using both simulated and real transcriptome data. From this experience, they developed UnPaSt, an optimized method that can detect patterns in omic datasets even with a limited number of subtype-predictive biomarkers. The authors evaluated all 23 methods on real-world breast cancer and asthma transcriptomics data, showing that UnPaSt significantly outperforms its closest competitors in both test datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This study is about finding different groups of people who have the same disease. Right now, doctors often use a lot of tests to figure out which treatment will work best for each person. But what if there was a way to group people into smaller categories that would help doctors choose the right treatment? That’s what this research paper is trying to do. The scientists tried 22 different ways to group people based on their medical test results and found one method that works really well. This new method, called UnPaSt, can even work with limited information. It was tested on two different types of diseases: breast cancer and asthma. The results show that this new method is better than others at finding the right groups. |
Keywords
* Artificial intelligence * Unsupervised
