arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2103.13047cs.LGq-bio.QM
收藏

知识感知的对比分子图学习

Knowledge-aware Contrastive Molecular Graph Learning

College of Computer Science and Technology, Zhejiang University(浙江大学计算机科学与技术学院);Pharmaceutical Informatics Institute, College of Pharmaceutical Sciences, Zhejiang University(浙江大学药学院药物信息学研究所)

Yin Fang, Haihong Yang, Xiang Zhuang, Xin Shao, Xiaohui Fan, Huajun Chen

摘要:

在分子表示学习中利用指纹图谱和官能团等领域知识对于化学性质预测和药物发现至关重要。在隐式地建模图结构与分子性质之间的关系时,现有的工作很难捕捉结构或性质的变化以及复杂的结构,原子词汇量小得多,原子频率高。在本文中,我们提出了对比知识感知GNN(CKGNN)的自我监督分子表示学习融合领域知识到分子图表示。在对比学习框架下,通过知识感知的分子编码器对领域知识进行显式编码,确保生成的分子嵌入具有化学领域知识,能够区分化学式相似但功能不同的分子。在8个公开数据集上的实验表明,该模型的有效性,与强有力的竞争对手相比,平均有6%的绝对改进.消融研究和进一步的调查也验证了两全其美:将化学领域的知识纳入自监督学习。

英文摘要:

Leveraging domain knowledge including fingerprints and functional groups in molecular representation learning is crucial for chemical property prediction and drug discovery. When modeling the relation between graph structure and molecular properties implicitly, existing works can hardly capture structural or property changes and complex structure, with much smaller atom vocabulary and highly frequent atoms. In this paper, we propose the Contrastive Knowledge-aware GNN (CKGNN) for self-supervised molecular representation learning to fuse domain knowledge into molecular graph representation. We explicitly encode domain knowledge via knowledge-aware molecular encoder under the contrastive learning framework, ensuring that the generated molecular embeddings equipped with chemical domain knowledge to distinguish molecules with similar chemical formula but dissimilar functions. Extensive experiments on 8 public datasets demonstrate the effectiveness of our model with a 6\% absolute improvement on average against strong competitors. Ablation study and further investigation also verify the best of both worlds: incorporation of chemical domain knowledge into self-supervised learning.

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。

↑