MAGNET: Multi-view graph autoencoder with cell-gene attention for cell interaction network reconstruction from spatial transcriptomics
Chengxu Han, Zhangdi Song, Zimu Xu, Jiaxing Chen · 2025
Read MAGNET: Multi-view graph autoencoder with cell-gene attention for cell interaction network reconstruction from spatial transcriptomics by Chengxu Han, Zhangdi Song, Zimu Xu, Jiaxing Chen (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-12716704.txt.
Description
Chengxu Han and co-authors present a 2025 journal article on rebuilding cell interaction networks from spatial gene data. Tissue maps show where cells sit and which genes they run, but inferring who signals whom is hard. Past models merged outside signals and inside gene control only at the last step, by simple joining. The team's model, MAGNET, links the two throughout.
A cell-gene attention module ties each cell's surroundings to its inner gene activity in one representation. On three benchmark maps the model beat rivals, and removing the attention module cut its score sharply. Applied to breast cancer data, it split cancer cells into immune-evading and self-growing groups.
Catalog record
| CREATOR | Chengxu Han, Zhangdi Song, Zimu Xu, Jiaxing Chen |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 53 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |