Interpretable Machine Learning on Metabolomics Data Reveals Biomarkers for Parkinson's Disease.
Interpretable Machine Learning on Metabolomics Data Reveals Biomarkers for Parkinson's Disease.
复制标题
DOI：
10.1021/acscentsci.2c01468
复制
发表时间：
2023-05-24
影响因子：
18.2
通讯作者：
Donald, William A.
中科院分区：
文献类型：
--
作者：
Zhang, J. Diana;Xue, Chonghua;Kolachalama, Vijaya B.;Donald, William A.
机器学习（ML）与代谢组学的结合为疾病的早期诊断提供了机会。然而，ML的准确性和从代谢组学获得的信息的程度可能是有限的，这是由于与解释疾病预测模型和分析具有相关和“噪声”丰度的许多化学特征相关的挑战。在这里，我们报告了一个可解释的神经网络（NN）框架，可以准确预测疾病，并使用整个代谢组学数据集识别重要的生物标志物，而无需先验特征选择。NN方法从血浆代谢组学数据预测帕金森病（PD）的性能显著高于其他ML方法，平均曲线下面积>0.995。PD特异性标志物，早于临床PD诊断，并有助于早期疾病的预测显着确定包括外源性多氟烷基物质。预计这种准确和可解释的基于NN的方法可以使用代谢组学和其他非靶向组学方法改善许多疾病的诊断性能。对整个代谢组学数据集进行可解释的机器学习可以显着提高诊断性能，并回顾性地“挖掘”疾病生物标志物。
The use of machine learning (ML) with metabolomics provides opportunities for the early diagnosis of disease. However, the accuracy of ML and extent of information obtained from metabolomics can be limited owing to challenges associated with interpreting disease prediction models and analyzing many chemical features with abundances that are correlated and “noisy”. Here, we report an interpretable neural network (NN) framework to accurately predict disease and identify significant biomarkers using whole metabolomics data sets without a priori feature selection. The performance of the NN approach for predicting Parkinson’s disease (PD) from blood plasma metabolomics data is significantly higher than other ML methods with a mean area under the curve of >0.995. PD-specific markers that predate clinical PD diagnosis and contribute significantly to early disease prediction were identified including an exogenous polyfluoroalkyl substance. It is anticipated that this accurate and interpretable NN-based approach can improve diagnostic performance for many diseases using metabolomics and other untargeted ‘omics methods. Interpretable machine learning on whole metabolomics data sets can significantly improve diagnostic performance and retrospectively “mine” disease biomarkers.
登录
查看更多内容
影响因子：
10.4
作者：
Calafat, Antonia M;Wong, Lee-Yang;Kuklenyik, Zsuzsanna;Reidy, John A;Needham, Larry L
通讯作者：
Needham, Larry L
DOI：
10.1016/s1474-4422(17)30299-5
发表时间：
2017-11
期刊：
The Lancet. Neurology
影响因子：
--
作者：
GBD 2015 Neurological Disorders Collaborator Group
通讯作者：
GBD 2015 Neurological Disorders Collaborator Group
DOI：
10.1038/s41531-021-00216-4
发表时间：
2021-08-16
期刊：
NPJ Parkinson's disease
影响因子：
--
作者：
Gonzalez-Riano C;Saiz J;Barbas C;Bergareche A;Huerta JM;Ardanaz E;Konjevod M;Mondragon E;Erro ME;Chirlaque MD;Abilleira E;Goñi-Irigoyen F;Amiano P
通讯作者：
Amiano P
影响因子：
5.3
作者：
Hu, Haiyan;Zhou, Yuehan;Yan, Guangmei
通讯作者：
Yan, Guangmei
影响因子：
4.4
作者：
Chicco, Davide;Jurman, Giuseppe
通讯作者：
Jurman, Giuseppe