arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2201.04929cs.LG
收藏

改进基于VAE的分子表征方法预测化合物性质

Improving VAE based molecular representations for compound property prediction

YerevaNN;Institute of Molecular Biology, National Academy of Sciences of RA(亚美尼亚共和国国家科学院分子生物学研究所);University of Würzburg(维尔茨堡大学)

A. Tevosyan, L. Khondkaryan, H. Khachatrian, G. Tadevosyan, L. Apresyan, N. Babayan, H. Stopper, Z. Navoyan

摘要:

在化学信息学的许多重要任务中,收集标记数据是耗时的,并且需要昂贵的实验。近年来,机器学习已被用于使用大规模未标记的分子数据集来学习分子的丰富表示,并将知识转移到有限数据集上来解决更具挑战性的任务。变分自编码器是已经提出的用于执行化学性质预测和分子生成任务的转移的工具之一。在这项工作中,我们提出了一个简单的方法来提高机器学习模型的化学性质预测性能,通过将额外的信息相关的分子描述符的表示学习变分自编码器。在三个性能预测问题上对该方法进行了验证。我们探讨了合并的描述符的数量,描述符和目标属性之间的相关性,数据集的大小等的影响。最后,我们展示了属性预测模型的性能与属性预测数据集和表示空间中较大的未标记数据集之间的距离之间的关系。

英文摘要:

Collecting labeled data for many important tasks in chemoinformatics is time consuming and requires expensive experiments. In recent years, machine learning has been used to learn rich representations of molecules using large scale unlabeled molecular datasets and transfer the knowledge to solve the more challenging tasks with limited datasets. Variational autoencoders are one of the tools that have been proposed to perform the transfer for both chemical property prediction and molecular generation tasks. In this work we propose a simple method to improve chemical property prediction performance of machine learning models by incorporating additional information on correlated molecular descriptors in the representations learned by variational autoencoders. We verify the method on three property prediction asks. We explore the impact of the number of incorporated descriptors, correlation between the descriptors and the target properties, sizes of the datasets etc. Finally, we show the relation between the performance of property prediction models and the distance between property prediction dataset and the larger unlabeled dataset in the representation space.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑