School of Informatics , University of Skövde , 541 28 Skövde , Sweden.
Medicinal Chemistry, Early Cardiovascular, Renal and Metabolism, R&D BioPharmaceuticals , AstraZeneca , 431 83 Mölndal , Sweden.
J Chem Inf Model. 2019 Jul 22;59(7):3166-3176. doi: 10.1021/acs.jcim.9b00325. Epub 2019 Jul 5.
In medicinal chemistry programs it is key to design and make compounds that are efficacious and safe. This is a long, complex, and difficult multiparameter optimization process, often including several properties with orthogonal trends. New methods for the automated design of compounds against profiles of multiple properties are thus of great value. Here we present a fragment-based reinforcement learning approach based on an actor-critic model, for the generation of novel molecules with optimal properties. The actor and the critic are both modeled with bidirectional long short-term memory (LSTM) networks. The AI method learns how to generate new compounds with desired properties by starting from an initial set of lead molecules and then improving these by replacing some of their fragments. A balanced binary tree based on the similarity of fragments is used in the generative process to bias the output toward structurally similar molecules. The method is demonstrated by a case study showing that 93% of the generated molecules are chemically valid and more than a third satisfy the targeted objectives, while there were none in the initial set.
在药物化学课程中，设计和制造有效且安全的化合物是关键。这是一个漫长、复杂且困难的多参数优化过程，通常涉及具有正交趋势的多个性质。因此，针对多种性质的自动化合物设计的新方法具有重要价值。在这里，我们提出了一种基于演员-评论家模型的基于片段的强化学习方法，用于生成具有最佳性质的新型分子。演员和评论家都使用双向长短时记忆网络（LSTM）进行建模。人工智能方法通过从一组初始先导分子开始，并通过替换其中一些片段来改进这些分子，从而学习如何生成具有所需性质的新化合物。在生成过程中，基于片段相似性的平衡二叉树用于使输出偏向结构相似的分子。通过案例研究证明了该方法，其中 93%的生成分子在化学上是有效的，超过三分之一的分子满足目标，而初始集中没有一个分子。
