Small molecule generation via disentangled representation learning
Small molecule generation via disentangled representation learning
复制标题
DOI：
10.1093/bioinformatics/btac296
复制
发表时间：
2022-05
期刊：
影响因子：
5.8
通讯作者：
Yuanqi Du;Xiaojie Guo;Yinkai Wang;Amarda Shehu;Liang Zhao
中科院分区：
文献类型：
--
作者：
Yuanqi Du;Xiaojie Guo;Yinkai Wang;Amarda Shehu;Liang Zhao
动机扩展我们对小分子的了解，超越自然界已知的或在湿实验室中设计的，有望显着推进化学信息学，药物发现，生物技术和材料科学。计算机分子设计仍然具有挑战性，主要是由于化学空间的复杂性以及化学结构和生物学特性之间的非平凡关系。直接从数据中学习的深度生成模型很有趣，但它们还没有在学习的表示中表现出可解释性，因此我们可以更多地了解化学和生物空间之间的关系。在本文中，我们推进研究的解纠缠表示学习的小分子生成。我们建立在我们和其他人最近在深度图生成框架上的工作基础上，该框架通过基于图形的小分子表示来捕获原子相互作用。方法上的新奇是我们如何利用图变分自动编码器框架中的解纠缠概念来生成生物相关的小分子并增强模型的可解释性。结果与最先进的模型相比，广泛的定性和定量实验评估表明我们的解缠框架的优越性。我们相信这项工作是解决具有深度生成框架的小分子生成的关键挑战的重要一步。培训和生成的数据可在https://ieee-dataport.org/documents/dataset-disentangled-representation-learning-interpretable-molecule-generation上获得。所有代码都可以在https://anonymous.4open.science/r/D-MolVAE-2799/上获得。补充数据可在生物信息学在线获得。
MOTIVATION Expanding our knowledge of small molecules beyond what is known in nature or designed in wet laboratories promises to significantly advance cheminformatics, drug discovery, biotechnology, and material science. In-silico molecular design remains challenging, primarily due to the complexity of the chemical space and the non-trivial relationship between chemical structures and biological properties. Deep generative models that learn directly from data are intriguing, but they have yet to demonstrate interpretability in the learned representation, so we can learn more about the relationship between the chemical and biological space. In this paper, we advance research on disentangled representation learning for small molecule generation. We build on recent work by us and others on deep graph generative frameworks, which capture atomic interactions via a graph-based representation of a small molecule. The methodological novelty is how we leverage the concept of disentanglement in the graph variational autoencoder framework both to generate biologically-relevant small molecules and to enhance model interpretability. RESULTS Extensive qualitative and quantitative experimental evaluation in comparison with state of the art models demonstrate the superiority of our disentanglement framework. We believe this work is an important step to address key challenges in small molecule generation with deep generative frameworks. AVAILABILITY Training and generated data are made available at https://ieee-dataport.org/documents/dataset-disentangled-representation-learning-interpretable-molecule-generation. All code is made available at https://anonymous.4open.science/r/D-MolVAE-2799/. SUPPLEMENTARY INFORMATION Supplementary data are available at Bioinformatics online.