Choi Jonghwan, Seo Sangmin, Park Sanghyun
Department of Computer Science, Yonsei University, Yonsei-ro 50, 03722, Seoul, Republic of Korea.
UBLBio Corporation, Yeongtong-ro 237, 16679, Suwon, Gyeonggi-do, Republic of Korea.
J Cheminform. 2023 Jan 19;15(1):8. doi: 10.1186/s13321-023-00679-y.
Structure-constrained molecular generation is a promising approach to drug discovery. The goal of structure-constrained molecular generation is to produce a novel molecule that is similar to a given source molecule (e.g. hit molecules) but has enhanced chemical properties (for lead optimization). Many structure-constrained molecular generation models with superior performance in improving chemical properties have been proposed; however, they still have difficulty producing many novel molecules that satisfy both the high structural similarities to each source molecule and improved molecular properties.
We propose a structure-constrained molecular generation model that utilizes contractive and margin loss terms to simultaneously achieve property improvement and high structural similarity. The proposed model has two training phases; a generator first learns molecular representation vectors using metric learning with contractive and margin losses and then explores optimized molecular structure for target property improvement via reinforcement learning.
We demonstrate the superiority of our proposed method by comparing it with various state-of-the-art baselines and through ablation studies. Furthermore, we demonstrate the use of our method in drug discovery using an example of sorafenib-like molecular generation in patients with drug resistance.
结构受限的分子生成是一种很有前景的药物发现方法。结构受限的分子生成的目标是生成一种与给定源分子（例如先导化合物分子）相似但具有增强化学性质（用于先导化合物优化）的新型分子。已经提出了许多在改善化学性质方面具有卓越性能的结构受限分子生成模型；然而，它们在生成许多既与每个源分子具有高结构相似性又具有改善的分子性质的新型分子方面仍然存在困难。
我们提出了一种结构受限的分子生成模型，该模型利用收缩损失项和边界损失项来同时实现性质改善和高结构相似性。所提出的模型有两个训练阶段；生成器首先使用带有收缩损失和边界损失的度量学习来学习分子表示向量，然后通过强化学习探索用于改善目标性质的优化分子结构。
我们通过将我们提出的方法与各种最先进的基线方法进行比较并通过消融研究来证明我们方法的优越性。此外，我们以耐药患者中索拉非尼样分子生成的例子展示了我们的方法在药物发现中的应用。
