Deep Generative and Graph-Based Representation Learning for Multiomics Survival Stratification in Ovarian Cancer: Secondary Analysis
Carlos Marino, Claudia Diaz Paz, Qianqian Song · 2026
Read Deep Generative and Graph-Based Representation Learning for Multiomics Survival Stratification in Ovarian Cancer: Secondary Analysis by Carlos Marino, Claudia Diaz Paz, Qianqian Song (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/PMC13592661.txt.
Description
Carlos Marino and two colleagues sort ovarian cancer patients by survival outlook using deep learning on multiomics data. Tumor profiles spanning genes, gene activity, and chemical tags are rich but huge and messy, and the patient pool is small. Their pipeline compresses each patient's profile with a variational autoencoder, then clusters the compressed forms and tests whether the groups live differently. A graph based step is described but was not tested here.
After record matching fixes, 291 patients remained. Two clusters emerged as the cleanest split, and their survival curves parted clearly. One group carried about half the death risk of the other. Marino and the team say the learned profiles hold real survival signal, while outside validation and the graph step remain future work.
Catalog record
| CREATOR | Carlos Marino, Claudia Diaz Paz, Qianqian Song |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 32 KB |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |