CMA-DTI: a cross-modal fusion and attentive interaction network for interpretable drug-target interaction prediction
Chi Qin, Denggao Zheng, Ziyang Wang, Yu Li, Jingrui Cao, Houchun Qiu, Hongxing Kan, Liping Sun, Yu Liu, Jili Hu · 2026
Read CMA-DTI: a cross-modal fusion and attentive interaction network for interpretable drug-target interaction prediction by Chi Qin, Denggao Zheng, Ziyang Wang, Yu Li, Jingrui Cao, Houchun Qiu, Hongxing Kan, Liping Sun, Yu Liu, Jili Hu (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-13294260.txt.
Description
Chi Qin and co-authors put forth CMA-DTI, a network that forecasts if a drug will interact with a target protein. Their 2026 piece deals with an early step in drug discovery. At that step, forecasts that can be read and checked are still hard to make.
The frame joins data shown as molecular graphs, data shown as text-based chemical forms, and protein residue embeddings from models trained in advance. Cross-attention links graph nodes with chemical tokens in the drug data. A second attention step picks up local links between drug traits and protein residues. On two benchmark datasets, the model did as well as set methods it was set next to. It also kept its power for drugs and targets it had not seen. And attention picked out residues close to known binding pockets.
Catalog record
| CREATOR | Chi Qin, Denggao Zheng, Ziyang Wang, Yu Li, Jingrui Cao, Houchun Qiu, Hongxing Kan, Liping Sun, Yu Liu, Jili Hu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 51 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |