Zhiqiang Zhong
ABSTRACT
Conventional Artificial Intelligence models are heavily limited in handling complex biomedical structures (such as 2D or 3D protein and molecule structures) and providing interpretations for outputs, which hinders their practical application. Graph Machine Learning (GML) has gained considerable attention for its exceptional ability to model graph-structured biomedical data and investigate their properties and functional relationships. Despite extensive efforts, GML methods still suffer from several deficiencies, such as the limited ability to handle supervision sparsity and provide interpretability in learning and inference processes and their ineffectiveness in utilising relevant domain knowledge. In response, recent studies have proposed integrating external biomedical knowledge into the GML pipeline to realise more precise and interpretable drug discovery with limited training instances.
This tutorial presents a comprehensive overview of long-standing drug discovery principles, provides the foundational concepts and cutting-edge techniques for graph-structured data and knowledge databases, and formally summarises Knowledge-augmented Graph Machine Learning (KaGML) for drug discovery. We have recently completed a survey of KaGML works that organises the outstanding approaches into four categories following a novel-defined taxonomy. This tutorial will present the result of this scholarly work. To encourage audience participation and facilitate research in this promptly emerging field, we also share valuable practical resources for intelligent drug discovery and provide an in-depth discussion of the potential avenues for future advancements.
- Simon Batzner, Albert Musaelian, Lixin Sun, et al. 2022. E (3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials. Nat. Commun. , Vol. 13, 1 (2022), 1--11.Google Scholar
Cross Ref
- Roy Benjamin, Uriel Singer, and Kira Radinsky. 2022. Graph Neural Networks Pretraining Through Inherent Supervision for Molecular Property Prediction. In CIKM. ACM, 2903--2912.Google Scholar
- Yin Fang, Qiang Zhang, Haihong Yang, et al. 2022. Molecular Contrastive Learning with Chemical Element Knowledge Graph. In AAAI. 3968--3976.Google Scholar
- Yun Hao, Joseph D. Romano, and Jason H. Moore. 2022. Knowledge-guided deep learning models of drug toxicity improve interpretation. Patterns, Vol. 3, 9 (2022), 100565.Google ScholarCross Ref
- Tatsuya Hasebe. 2021. Knowledge-Embedded Message-Passing Neural Networks: Improving Molecular Property Prediction with Human Knowledge. ACS Omega, Vol. 6, 42 (2021), 27955--27967.Google ScholarCross Ref
- Zewei Ji, Runhan Shi, Jiarui Lu, et al. 2022. ReLMole: Molecular Representation Learning Based on Two-Level Graph Similarities. J. Chem. Inf. Model. (2022).Google Scholar
- John Jumper, Richard Evans, Alexander Pritzel, et al. 2021. Highly accurate protein structure prediction with AlphaFold. Nature, Vol. 596, 7873 (2021), 583--589.Google Scholar
- Han Li, Dan Zhao, and Jianyang Zeng. 2022. KPGT: Knowledge-Guided Pre-training of Graph Transformer for Molecular Property Prediction. In KDD. ACM, 857--867.Google ScholarDigital Library
- Xuan Lin, Zhe Quan, Zhi-Jie Wang, et al. 2020. KGNN: Knowledge Graph Neural Network for Drug-Drug Interaction Prediction. In IJCAI. 2739--2745.Google Scholar
- Shengchao Liu, Meng Qu, Zuobai Zhang, et al. 2022a. Structured multi-task learning for molecular property prediction. In AISTATS, Vol. 151. PMLR, 8906--8920.Google Scholar
- Shengchao Liu, Hanchen Wang, Weiyang Liu, et al. 2022b. Pre-training Molecular Graph Representation with 3D Geometry. In ICLR.Google Scholar
- Wan Xiang Shen, Xian Zeng, Feng Zhu, et al. 2021. Out-of-the-box deep learning prediction of pharmaceutical properties by broadly learned knowledge-based molecular representations. Nat. Mach. Intell, Vol. 3, 4 (2021), 334--343.Google ScholarCross Ref
- Jihye Shin, Yinhua Piao, Dongmin Bang, et al. 2022. DRPreter: Interpretable Anticancer Drug Response Prediction Using Knowledge-Guided Graph Neural Networks and Transformer. Int. J. Mol. Sci. , Vol. 23, 22 (2022), 13919.Google ScholarCross Ref
- Mengying Sun, Jing Xing, Huijun Wang, et al. 2021. MoCL: Data-driven Molecular Fingerprint via Knowledge-aware Contrastive Learning from Molecular Graph. In KDD. ACM, 3585--3594.Google Scholar
- Yuancheng Sun, Yimeng Chen, Weizhi Ma, et al. 2022. PEMP: Leveraging Physics Properties to Enhance Molecular Property Prediction. In CIKM. ACM, 3505--3513.Google Scholar
- Kevin Yang, Kyle Swanson, Wengong Jin, et al. 2019. Analyzing learned molecular representations for property prediction. J. Chem. Inf. Model. , Vol. 59, 8 (2019), 3370--3388.Google ScholarCross Ref
- Yue Yu, Kexin Huang, Chao Zhang, et al. 2021. SumGNN: multi-typed drug interaction prediction via efficient knowledge graph summarization. Bioinform. , Vol. 37, 18 (2021), 2988--2995.Google ScholarCross Ref
- Ningyu Zhang, Zhen Bi, Xiaozhuan Liang, et al. 2022. OntoProtein: Protein Pretraining With Gene Ontology Embedding. In ICLR.Google Scholar
- Zhiqiang Zhong, Anastasia Barkova, and Davide Mottin. 2023. Knowledge-augmented Graph Machine Learning for Drug Discovery: A Survey from Precision to Interpretability. CoRR , Vol. abs/2302.08261 (2023).Google Scholar
Index Terms
- Knowledge-augmented Graph Machine Learning for Drug Discovery: From Precision to Interpretability
Recommendations
Artificial Intelligence for Drug Discovery
KDD '21: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data MiningDrug discovery is a long and costly process, taking on average 10 years and 2.5 billion dollars to develop a new drug. Artificial intelligence has the potential to significantly accelerate the process of drug discovery by analyzing a large amount of ...
Machine Learning on Mining Potential Adverse Drug Reactions for Pharmacovigilance
DSIT 2021: 2021 4th International Conference on Data Science and Information TechnologyDrugs are related to the lives and health of patients. Mining and discovering potential adverse drug reactions from a large number of unstructured texts play an important role in drug research and pharmacovigilance. It is a hot research issue in the ...
Machine Learning in Drug Discovery: A Review
AbstractThis review provides the feasible literature on drug discovery through ML tools and techniques that are enforced in every phase of drug development to accelerate the research process and deduce the risk and expenditure in clinical trials. Machine ...
Access Granted
Comments