DeepEnhancerPPO: An Interpretable Deep Learning Approach for Enhancer Classification
Xuechen Mu, Zhenyu Huang, Qiufen Chen, Bocheng Shi, Long Xu, Ying Xu, Kai Zhang · 2024
Read DeepEnhancerPPO: An Interpretable Deep Learning Approach for Enhancer Classification by Xuechen Mu, Zhenyu Huang, Qiufen Chen, Bocheng Shi, Long Xu, Ying Xu, Kai Zhang (2024) 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-11641363.txt.
Description
Enhancers are short DNA segments that dial gene activity up or down, and sorting them by type and strength helps decode the genome's control system. Mu and six co-authors build DeepEnhancerPPO, a deep learning model for that sorting job. It blends a residual network and a transformer to catch local patterns and long-range context at once. A reinforcement learning step then trims the combined features down to the most useful ones.
In tests, each part of the design helped, with the residual network and the learning-based trimming helping most. The model beat rivals on fresh data, topping the runner-up by 6.7 percent in accuracy for enhancer categories. For strength classes it stayed near the top among twenty-five models. The trimming step also makes the decisions easier to interpret. The authors offer the tool as a robust aid for genome annotation.
Catalog record
| CREATOR | Xuechen Mu, Zhenyu Huang, Qiufen Chen, Bocheng Shi, Long Xu, Ying Xu, Kai Zhang |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |