Interpretable multi-omics integration across mixed-order tensors with MANTRA
Kevin De Azevedo, Yusuf Berk Oruc, Florian Buettner · 2026
Read Interpretable multi-omics integration across mixed-order tensors with MANTRA by Kevin De Azevedo, Yusuf Berk Oruc, Florian Buettner (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/PMC13538497.txt.
Description
Kevin De Azevedo, Yusuf Berk Oruc, and Florian Buettner present MANTRA, a framework for merging molecular data tables of mixed shapes in a 2026 methods study. Modern designs pair cubes such as patient by drug by condition with flat gene tables, breaking standard mergers. The probabilistic design learns shared hidden factors, respects higher-order structure, and fills missing entries in a principled way. Two leukemia data sets supplied tests.
Joint analysis of drug response cubes with gene tables surfaced patient subgroups that single-view and flat methods missed. A single-cell multi-layer study revealed a fresh subgroup defined by a distinct program in a rare immune cell type, tying disease variety to one cell source. The authors offer the tool for readable mining of tangled trial and single-cell data.
Catalog record
| CREATOR | Kevin De Azevedo, Yusuf Berk Oruc, Florian Buettner |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 48722 |
| LICENSE | https://creativecommons.org/publicdomain/zero/1.0 |
| RIGHTS | CC0 |
| SOURCE | Europe PMC |