De Novo Design of Molecules Towards Biased Properties via a Deep Generative Framework and Iterative Transfer Learning
De Novo Design of Molecules Towards Biased Properties via a Deep Generative Framework and Iterative Transfer Learning
复制标题
DOI：
--
复制
发表时间：
2023
影响因子：
8.4
通讯作者：
Kianoosh Sattari;Dawei Li;Yunchao Xie;O. Isayev;Jian Lin
中科院分区：
文献类型：
--
作者：
Kianoosh Sattari;Dawei Li;Yunchao Xie;O. Isayev;Jian Lin
具有靶向性质的分子的从头设计代表了分子开发的新前沿。尽管取得了巨大进展，但仍然存在两个主要挑战，即，（i）产生具有靶向和可量化性质的新分子;（ii）产生具有超出训练数据集中的范围的性质值的分子。为了应对这些挑战，我们提出了一种新的增强回归和条件生成对抗网络（RRCGAN），以生成具有目标热容（CV）值的化学有效的小分子作为概念验证研究。如通过DFT验证的，约80%的生成的样品具有< 20%的目标CV值的相对误差（RE）。为了使具有超出原始训练分子的范围的Cv值的分子的生成偏置，应用迁移学习来迭代地重新训练RRCGAN模型。经过两次迁移学习迭代后，生成分子的平均CV从初始训练数据集中显示的平均值31.6 cal/（mol·K）增加到44.0 cal/（mol·K）。这表明计算方法铺平了一条新的途径，发现小分子与偏见的性质。
De Novo design of molecules with targeted properties represents a new frontier in molecule development. Despite enormous progress, two main challenges remain, i.e., ( i ) generation of novel molecules with targeted and quantifiable properties; ( ii ) generated molecules having property values beyond the range in the training dataset. To tackle these challenges, we propose a novel reinforced regressional and conditional generative adversarial network (RRCGAN) to generate chemically valid, small molecules with targeted heat capacity ( C v ) values as a proof-of-concept study. As validated by DFT, ~80% of the generated samples have a relative error (RE) of < 20% of the targeted C v values. To bias the generation of molecules with the C v values beyond the range of the original training molecules, transfer learning was applied to iteratively retrain the RRCGAN model. After only two iterations of transfer learning, the mean C v of the generated molecules increases to 44.0 cal/(mol·K) from the mean value of 31.6 cal/(mol·K) shown in the initial training dataset. This demonstrated computation methodology paves a new avenue to discovering small molecules with biased properties.