Explainable Fragment-based Molecular Property Attribution

Lingxiang Jia, Zunlei Feng, Haotian Zhang, Jie Song, Zipeng Zhong, Shaolun Yao, Mingli Song

Abstract

“AI & Drug Discovery” mode has significantly promoted drug development and achieved excellent performance, especially with the rapid development of deep learning, making remarkable contributions to protecting human physiological health. However, due to the “black-box” characteristic of the deep learning model, the decision route and predicted results in different research stages assisted by deep models are usually unexplainable, limiting their application in practice and more in-depth research of drug discovery. Focusing on the drug molecules, we propose an explainable fragment-based molecular property attribution technique for analyzing the influence of particular molecule fragments on properties and the relationship between the molecular properties in this paper. Quantitative experiments on 42 benchmark property tasks demonstrate that 325 attribution fragments, which account for 90% of the overall attribution results obtained by the proposed method, have positive relevance to the corresponding property tasks. More impressively, most of the attribution results randomly selected are consistent with the existing mechanism explanations. The discovery mentioned above provides a reference standard for assisting researchers in developing more specific and practical drug molecule studies, such as synthesizing molecular with the targeted property using a fragment obtained from the attribution method. Corresponding author(s) Email: [email protected]
BibTeX
@inproceedings{jia2022explainable,
  author = {Lingxiang Jia and Zunlei Feng and Haotian Zhang and Jie Song and Zipeng Zhong and Shaolun Yao and Mingli Song},
  title = {{Explainable Fragment-based Molecular Property Attribution}},
  booktitle = {Adv. Intell. Syst},
  year = {2022},
  url = {https://openreview.net/forum?id=phOkIjTo_f5},
}

Loading related content...

Figure