DAGFormer: A graph-based domain adaptation approach for single-cell cancer drug response prediction
Fen Yan, ZhiHua Du, Yu-An Huang, Pedro Mendes, Wei Li, Pedro Mendes, Wei Li, Pedro Mendes, Wei Li · 2025
Read DAGFormer: A graph-based domain adaptation approach for single-cell cancer drug response prediction by Fen Yan, ZhiHua Du, Yu-An Huang, Pedro Mendes, Wei Li, Pedro Mendes, Wei Li, Pedro Mendes, Wei Li (2025) 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-12795466.txt.
Description
Fen Yan, ZhiHua Du, and Yu-An Huang present a prediction framework in 2025 for how single cancer cells respond to drugs. Bulk gene reading averages millions of cells and hides the differences that drive drug resistance, while single-cell reading is costly and small-scale. Joining the two data types is hampered by batch shifts and by models that treat cells as isolated.
Their framework builds neighbor graphs among cells and adapts knowledge across the bulk and single-cell domains at graph level. A two-part decoder keeps shared signals apart from source-specific ones. Benchmarks on ten independent single-cell data sets show it beating existing methods, which the authors read as proof of robust, accurate prediction.
Catalog record
| CREATOR | Fen Yan, ZhiHua Du, Yu-An Huang, Pedro Mendes, Wei Li, Pedro Mendes, Wei Li, Pedro Mendes, Wei Li |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 101 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |