STED: flexible cross-modal topic modeling infers cell-type-specific regulatory landscapes from bulk epigenomics
Yunxi Liao, Dongyu Zhao · 2026
Read STED: flexible cross-modal topic modeling infers cell-type-specific regulatory landscapes from bulk epigenomics by Yunxi Liao, Dongyu Zhao (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/pmc-13317755.txt.
Description
Yunxi Liao and Dongyu Zhao present STED, a tool that reads cell-type signals out of mixed tissue data, in a 2026 paper. Tissues mix many cell types, and single-cell methods that split them are sparse, costly, and slow. Vast stores of mixed, or bulk, gene data already exist. STED uses single-cell RNA references to unpack the bulk record into cell-type parts.
Its topic modeling core can run in a classic mode or borrow a language-model method for rich cell states. A gene activity score then bridges from RNA topics to DNA access and mark data. Tests on blood, mouse brain, and fish ear data beat older rigid tools, even across platforms. In a cancer case, STED found tumor enhancers hidden in the bulk signal.
Catalog record
| CREATOR | Yunxi Liao, Dongyu Zhao |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 55 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |