Drug discovery and development are expensive and time-consuming processes. Developing a new drug from discovery to regulatory approval may take 12 years and cost up to $2.8 billion. Furthermore, there is a high failure rate (1:5000) at each stage of drug development [2]. With the remarkable success of machine learning in various application fields, we are seeing increasing interest in the application of machine learning in drug discovery and development [1-6]. Here, we focus on this topic and review the literature on deep learning in drug discovery.
Hosein Fooladi divides the application of deep learning in drug discovery mainly into three different categories: Drug properties prediction, De Novo drug design, and Drug-target interaction (DTI) prediction [1]. The 29th IJCAI discusses key classes of methods for tackling these drug-related tasks: Generative models, Reinforcement learning, and Deep representation learning [2].
[1] Fooladi, H. (2018, October 31). Review: Deep Learning In Drug Discovery. Hosein Fooladi. https://hfooladi.github.io/posts/2018/10/Review-Deep-Learning-In-Drug-Discovery/.
[2] Machine Learning for Drug Development Tutorial at the 29th International Joint Conference on Artificial Intelligence (IJCAI2020). https://zitniklab.hms.harvard.edu/drugml/.
[3] deepakvraghavan. (2018, May 15) Real World Deep Neural Network Architectures for Pharma Industry. https://deepakvraghavan.medium.com/real-world-deep-neural-network-architectures-for-pharma-industry-a6e885f8038f/.
[4] Schneider, G. (2018). Automating drug discovery. Nature reviews drug discovery, 17(2), 97-113.
[5] Neil Savage. (2021, May 27). Tapping into the drug discovery potential of AI. https://www.nature.com/articles/d43747-021-00045-7.
[6] 唐巧. (2020, Oct 24) AI+新药领域行业发展. https://vcbeat.top/48626.
[1] A Deep Learning Library for Compound and Protein Modeling DTI, Drug Property, PPI, DDI, Protein Function Prediction: https://github.com/kexinhuang12345/DeepPurpose.
[2] A Deep Learning based Efficacy Prediction System for Drug Discovery: https://github.com/kekegg/DLEPS.
[3] Classification of Drug Like molecules using Artificial Neural Network. https://gananath.github.io/drugai.html; https://github.com/Gananath/DrugAI,
The following is an example, please download for details: https://github.com/ugggddd/DrugDiscoveryAI/blob/master/Drug_AI_Literature_Review.xlsx. (Ongoing).
Year | Title | Author | Organization | Journal | IF | Citation | DOI | DATA | Code |
---|---|---|---|---|---|---|---|---|---|
2021 | Could graph neural networks learn better molecular representation for drug discovery? A comparison study of descriptor-based and graph-based models | Guangyong Chen | Journal of cheminformatics | Shenzhen Institute of Advanced Technology, Chinese Academy of Science | 5.514 (Q2) | 5 | https://doi.org/10.1186/s13321-020-00479-8 | ||
2020 | Machine learning approaches to drug response prediction: challenges and recent progress | George Adam; Anna Goldenberg | University of Minnesota | npj Precision Oncology | 8.25 | 30 | https://doi.org/10.1038/s41698-020-0122-1 |
Author Name | Organization | Profile |
---|---|---|
Jimeng Sun | UIUC | http://www.sunlab.org/ |
Hosein Fooladi | Sharif University | https://hfooladi.github.io/ |
Kexin Huang | Harvard/Stanford | https://www.kexinhuang.com/ |
Cao (Danica) Xiao | Director of Machine Learning at IQVIA | https://scholar.google.com/citations?user=ahaV25EAAAAJ&hl=en |
计算机辅助药物设计中心(袁曙光教授课题组) | Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences | http://cadd.siat.ac.cn/home/ |
Guangyong Chen | Shenzhen Institute of Advanced Technology, Chinese Academy of Science | https://guangyongchen.github.io/ |
Zhaoping Xiong | ShanghaiTech University | https://scholar.google.com/citations?user=XZ8wFwkAAAAJ&hl=en |
Qingpeng Zhang | City University of Hong Kong | http://www.cityu.edu.hk/stfprofile/zhang.html |
IQVIA | Company | https://www.iqvia.com/ |
Recursion Pharmaceuticals | Company | https://www.recursion.com/ |
晶泰科技 | Company | https://www.jingtaikeji.com/zh-hans/ |
百图生科CRMBioMap | Company | http://www.biomap.com/ |
腾讯量子实验室Tencent Quantum Lab | Company | https://quantum.tencent.com/about/ |
云深智药 | Company | https://drug.ai.tencent.com/en |
DrugAI | 公众号 | https://www.zhihu.com/column/c_1155516810005778432 |