Shedding light on “Black Box” machine learning models for predicting the reactivity of HO radicals toward organic compounds

均方误差 虚假关系 数量结构-活动关系 人工神经网络 平均绝对误差 人工智能 支持向量机 Boosting(机器学习) 梯度升压 模式识别(心理学) 反应性(心理学) 机器学习 分子描述符 集成学习 数学 生物系统 计算机科学 统计 随机森林 病理 生物 医学 替代医学
作者
Shifa Zhong,Kai Zhang,Dong Wang,Huichun Zhang
出处
期刊:Chemical Engineering Journal [Elsevier BV]
卷期号:405: 126627-126627 被引量:97
标识
DOI:10.1016/j.cej.2020.126627
摘要

Developing quantitative structure-activity relationships (QSARs) is an important approach to predicting the reactivity of HO radicals toward newly emerged organic compounds. As compared with molecular descriptors-based and the group contribution method-based QSARs, a combined molecular fingerprint-machine learning (ML) method can more quickly and accurately develop such models for a growing number of contaminants. However, it is yet unknown whether this method makes predictions by choosing meaningful structural features rather than spurious ones, which is vital for trusting the models. In this study, we developed QSAR models for the logkHO values of 1089 organic compounds in the aqueous phase by two ML algorithms—deep neural networks (DNN) and eXtreme Gradient Boosting (XGBoost), and interpreted the built models by the SHapley Additive exPlanations (SHAP) method. The results showed that for the contribution of a given structural feature to logkHO for different compounds, DNN and XGBoost treated it as a fixed and variable value, respectively. We then developed an ensemble model combining the DNN with XGBoost, which achieved satisfactory predictive performance for all three datasets: Training dataset: R-square (R2) 0.89–0.91, root-mean-squared-error (RMSE) 0.21–0.23, and mean absolute error (MAE) 0.15–0.17; Validation dataset: R2 0.63–0.78, RMSE 0.29–0.32, and MAE 0.21–0.25; and Test dataset: R2 0.60–0.71, RMSE 0.30–0.35, and MAE 0.23–0.25. The SHAP method was further used to unveil that this ensemble model made predictions on logkHO based on a correct ‘understanding’ of the impact of electron-withdrawing and -donating groups and of the reactive sites in the compounds that can be attacked by HO. This study offered some much-needed mechanistic insights into a ML-assisted environmental task, which are important for evaluating the trustworthiness of the ML-based models, further improving the models for specific applications, and leveraging the implicit knowledge the models carry.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
飞云完成签到,获得积分10
刚刚
cppcppsmida完成签到,获得积分10
1秒前
哈ha完成签到,获得积分10
2秒前
jac1完成签到,获得积分10
3秒前
Rosaline发布了新的文献求助20
3秒前
下午好完成签到 ,获得积分10
4秒前
liu123456完成签到,获得积分10
4秒前
zj完成签到,获得积分10
4秒前
琴楼完成签到,获得积分10
4秒前
慕新完成签到,获得积分10
5秒前
6秒前
熄熄完成签到 ,获得积分10
6秒前
7秒前
梦XING完成签到 ,获得积分10
9秒前
9秒前
微笑的兔子完成签到,获得积分10
9秒前
hhhhmmmn完成签到,获得积分10
10秒前
12秒前
12秒前
小东同志完成签到,获得积分10
13秒前
Victoria完成签到,获得积分10
15秒前
沉默的寻凝完成签到 ,获得积分10
15秒前
建业完成签到,获得积分10
17秒前
18秒前
希雅完成签到 ,获得积分10
19秒前
ET完成签到,获得积分10
19秒前
rainy发布了新的文献求助30
19秒前
jiayueiyang完成签到,获得积分10
23秒前
liu bo完成签到,获得积分10
23秒前
Akim应助机灵的丸子采纳,获得10
23秒前
23秒前
翟淑雨发布了新的文献求助10
24秒前
科研小白完成签到,获得积分10
24秒前
25秒前
英姑应助建业采纳,获得10
27秒前
喜乐发布了新的文献求助30
27秒前
LL完成签到,获得积分10
27秒前
ynlqjqx发布了新的文献求助10
28秒前
温大善人完成签到,获得积分10
31秒前
back you up关注了科研通微信公众号
32秒前
高分求助中
Production Logging: Theoretical and Interpretive Elements 2700
Neuromuscular and Electrodiagnostic Medicine Board Review 1000
こんなに痛いのにどうして「なんでもない」と医者にいわれてしまうのでしょうか 510
The First Nuclear Era: The Life and Times of a Technological Fixer 500
Unusual formation of 4-diazo-3-nitriminopyrazoles upon acid nitration of pyrazolo[3,4-d][1,2,3]triazoles 500
岡本唐貴自伝的回想画集 500
Distinct Aggregation Behaviors and Rheological Responses of Two Terminally Functionalized Polyisoprenes with Different Quadruple Hydrogen Bonding Motifs 450
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 物理 生物化学 纳米技术 计算机科学 化学工程 内科学 复合材料 物理化学 电极 遗传学 量子力学 基因 冶金 催化作用
热门帖子
关注 科研通微信公众号,转发送积分 3671598
求助须知:如何正确求助?哪些是违规求助? 3228309
关于积分的说明 9779385
捐赠科研通 2938622
什么是DOI,文献DOI怎么找? 1610143
邀请新用户注册赠送积分活动 760547
科研通“疑难数据库(出版商)”最低求助积分说明 736093