HyperAttentionDTI: improving drug–protein interaction prediction by sequence-based deep learning with attention mechanism
Problem
Identifying drug-target interactions experimentally is slow and costly; accurate in-silico prediction could shorten drug development and repurposing timelines.
Why it matters
Drug discovery is expensive and time-consuming, and better interaction prediction could cut both.
What the AI did
HyperAttentionDTI, an end-to-end CNN with an attention mechanism, learns drug and protein sequence features and models atom–amino-acid interactions to predict interactions.
Results
not stated in source
Impact
Proposes a method claimed to improve DTI prediction, but no measured outcome is stated in the abstract.
Record
| Venue | Bioinformatics |
|---|---|
| Published | 2021 |
| Work type | article |
| Authors | Qichang Zhao, Haochen Zhao, Kai Zheng, Jianxin Wang |
| Citations | 274 (OpenAlex, at import time) |
| References | 30 |
| Model or method | HyperAttentionDTI (CNN with attention mechanism) |
| Organization | not stated in source |
| Primary source | https://doi.org/10.1093/bioinformatics/btab715 |
| Open access | closed |
| Keywords | drug-target interaction, deep learning, attention mechanism, CNN, drug repurposing |
Importance rating
significant — thousands to millions affected
Abstract states DTI identification is crucial to drug repurposing/discovery and accurate in-silico prediction can significantly shorten development time and reduce costs.