hERGAT

Title: hERGAT: Predicting hERG blockers using graph attention mechanism through atom- and molecule-level interaction analyses
Authors: Dohyeon Lee, Sunyong Yoo

Description

We present hERGAT, an interpretable deep learning model that utilizes graph attention to predict hERG blockers.

In addition, we provide a Python file that can be used to generate predictions from the model we trained.

If you want, you can train and predict new datasets from the structure of the model we proposed.

• Data
• hERG source code
• Atom- and molecule-level interaction analysis
• Figure used in this paper

Dependency

Python == 3.9.18 tensorflow == 2.15.0 keras == 2.15.0 scikit-learn==1.3.2 rdkit==2023.9.2

Contacts

If you have any questions or comments, please feel free to create an issue on github here, or email us:

• ldhyun7222@naver.com
• syyoo@jnu.ac.kr