CapsCarcino: A novel sparse data deep learning tool for predicting carcinogens.

CapsCarcino: A novel sparse data deep learning tool for predicting carcinogens.
复制标题

CapsCarcino:一种用于预测致癌物的新型稀疏数据深度学习工具。

DOI:
10.1016/j.fct.2019.110921
复制
发表时间:
2019
期刊:
Food Chem Toxicol
影响因子:
--
通讯作者:
Sheng-Yong Yang
Sheng-Yong Yang
中科院分区:
其他
文献类型:
--
作者:
Yi-Wei Wang;Lei Huang;Si-Wen Jiang;Kan Li;Jun Zou';Sheng-Yong Yang

文献摘要

参考文献

被引文献

相似文献

在药物发现的早期阶段确定化学致癌性对于防止致癌物对人类健康的不利影响至关重要。最近,人们对开发预测化学致癌性的计算方法产生了浓厚的兴趣。然而,许多现有方法的预测能力有限,并且还有很大的改进空间。在这里,我们开发了一种新的深度学习架构,称为 CapsCarcino,用于区分致癌物和非致癌物。 CapsCarcino基于动态路由算法构建,需要更少的数据,提取更全面的信息,并且不需要特征选择。我们发现 CapsCarcino 提供了显着提高的预测和泛化能力,并且优于其他五种机器学习模型。具体来说,CapsCarcino 的最佳模型在外部验证数据集上的准确率达到 85.0%。此外,我们发现 CapsCarcino 相对于其他方法的增强预测能力是稳健的,并且可以使用稀疏数据集来实现。仅使用 20% 的数据集进行训练,CapsCarcino 的性能与基于完整训练数据集的其他方法相当。进一步的机制分析表明,即使结构警报没有得到充分体现,CapsCarcino 也可以有效地了解致癌物的特征。结果表明CapsCarcino应该有助于致癌物风险评估。
Determining chemical carcinogenicity in the early stages of drug discovery is fundamentally important to prevent the adverse effect of carcinogens on human health. There has been a recent surge of interest in developing computational approaches to predict chemical carcinogenicity. However, the predictive power of many existing approaches is limited, and there is plenty of room for improvement. Here, we develop a new deep learning architecture, termed CapsCarcino, to distinguish between carcinogens and noncarcinogens. CapsCarcino is constructed based on a dynamic routing algorithm that requires less data, extracts more comprehensive information, and does not require feature selection. We find that CapsCarcino provides a significantly improved predictive and generalization ability over, and outperforms five other machine learning models. Specifically, the best model of CapsCarcino achieves an accuracy of 85.0% on an external validation dataset. In addition, we discover that the enhanced predictive capability of CapsCarcino over that of the other methods is robust and can be achieved using sparse datasets. Training on merely 20% of the dataset, CapsCarcino performs comparably to the other methods based on the full training dataset. Further mechanism analysis indicates that CapsCarcino could efficiently learn the characteristics of carcinogens even if structural alerts are insufficiently represented. The results indicate that CapsCarcino should be helpful for carcinogen risk assessment.
DOI: 10.1093/bioinformatics/btx806
发表时间: 2018-05-01
期刊: Bioinformatics (Oxford, England)
影响因子: --
作者:
Preuer K;Lewis RPI;Hochreiter S;Bender A;Bulusu KC;Klambauer G
通讯作者: Klambauer G
CarcinoPred-EL:使用分子指纹和集成学习方法预测化学物质致癌性的新模型
DOI: 10.1038/s41598-017-02365-0
发表时间: 2017-05-18
期刊: Scientific reports
影响因子: 4.6
作者:
Zhang L;Ai H;Chen W;Yin Z;Hu H;Zhu J;Zhao J;Zhao Q;Liu H
通讯作者: Liu H
通过复发性神经网络生成聚焦的分子库来发现药物。
DOI: 10.1021/acscentsci.7b00512
发表时间: 2018-01-24
影响因子: 18.2
作者:
Segler MHS;Kogej T;Tyrchan C;Waller MP
通讯作者: Waller MP
DOI: 10.1016/j.fct.2016.09.005
发表时间: 2016-11
期刊: Food and chemical toxicology : an international journal published for the British Industrial Biological Research Association
影响因子: --
作者:
Hui Zhang;Z. Cao;Meng Li;Yu-Zhi Li;Cheng Peng
通讯作者: Hui Zhang;Z. Cao;Meng Li;Yu-Zhi Li;Cheng Peng
DOI: 10.1289/ehp.95103680
发表时间: 1995-07
影响因子: 10.4
作者:
Fung VA;Barrett JC;Huff J
通讯作者: Huff J