Interpretation of chemical reaction yields with graph neural additive network

Prediction of chemical yields is crucial for exploring untapped chemical reactions and optimizing synthetic pathways for targeted compounds. Recently, graph neural networks have proven successful in achieving high predictive accuracy. However, they remain intrinsically black-box models, offering lim...

Full description

Saved in:
Bibliographic Details
Main Authors: Youngchun Kwon, Yongsik Jung, Youn-Suk Choi, Seokho Kang
Format: Article
Language:English
Published: IOP Publishing 2025-01-01
Series:Machine Learning: Science and Technology
Subjects:
Online Access:https://doi.org/10.1088/2632-2153/addfaa
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849723311279308800
author Youngchun Kwon
Yongsik Jung
Youn-Suk Choi
Seokho Kang
author_facet Youngchun Kwon
Yongsik Jung
Youn-Suk Choi
Seokho Kang
author_sort Youngchun Kwon
collection DOAJ
description Prediction of chemical yields is crucial for exploring untapped chemical reactions and optimizing synthetic pathways for targeted compounds. Recently, graph neural networks have proven successful in achieving high predictive accuracy. However, they remain intrinsically black-box models, offering limited interpretability. Understanding how each reaction component contributes to the yield of a chemical reaction can help identify critical factors driving the success or failure of reactions, thereby potentially revealing opportunities for yield optimization. In this study, we present a novel method for interpretable chemical reaction yield prediction, which represents the yield of a chemical reaction as a simple summation of component-wise contributions from individual reaction components. To build an interpretable prediction model, we introduce a graph neural additive network architecture, wherein shared neural networks process individual reaction components in an input reaction while leveraging a reaction-level embedding to derive their respective contributions. The predicted yield is obtained by summing these component-wise contributions. The model is trained using a learning objective designed to effectively quantify the contributions of individual components by amplifying the influence of significant components and suppressing that of less influential components. The experimental results on benchmark datasets demonstrated that the proposed method achieved both high predictive accuracy and interpretability, making it suitable for practical use in synthetic pathway design for real-world applications.
format Article
id doaj-art-9a2dfe49ee194ef8a3ad83de6c03ea7a
institution DOAJ
issn 2632-2153
language English
publishDate 2025-01-01
publisher IOP Publishing
record_format Article
series Machine Learning: Science and Technology
spelling doaj-art-9a2dfe49ee194ef8a3ad83de6c03ea7a2025-08-20T03:11:03ZengIOP PublishingMachine Learning: Science and Technology2632-21532025-01-016202505410.1088/2632-2153/addfaaInterpretation of chemical reaction yields with graph neural additive networkYoungchun Kwon0https://orcid.org/0000-0001-7911-3670Yongsik Jung1https://orcid.org/0000-0003-4496-2351Youn-Suk Choi2https://orcid.org/0000-0001-7119-8788Seokho Kang3https://orcid.org/0000-0002-0960-0294Samsung Advanced Institute of Technology , Samsung Electronics Co. Ltd 130 Samsung-ro, Yeongtong-gu, Suwon 16678, Republic of KoreaSamsung Advanced Institute of Technology , Samsung Electronics Co. Ltd 130 Samsung-ro, Yeongtong-gu, Suwon 16678, Republic of KoreaSamsung Advanced Institute of Technology , Samsung Electronics Co. Ltd 130 Samsung-ro, Yeongtong-gu, Suwon 16678, Republic of KoreaDepartment of Industrial Engineering , Sungkyunkwan University, 2066 Seobu-ro, Jangan-gu, Suwon 16419, Republic of KoreaPrediction of chemical yields is crucial for exploring untapped chemical reactions and optimizing synthetic pathways for targeted compounds. Recently, graph neural networks have proven successful in achieving high predictive accuracy. However, they remain intrinsically black-box models, offering limited interpretability. Understanding how each reaction component contributes to the yield of a chemical reaction can help identify critical factors driving the success or failure of reactions, thereby potentially revealing opportunities for yield optimization. In this study, we present a novel method for interpretable chemical reaction yield prediction, which represents the yield of a chemical reaction as a simple summation of component-wise contributions from individual reaction components. To build an interpretable prediction model, we introduce a graph neural additive network architecture, wherein shared neural networks process individual reaction components in an input reaction while leveraging a reaction-level embedding to derive their respective contributions. The predicted yield is obtained by summing these component-wise contributions. The model is trained using a learning objective designed to effectively quantify the contributions of individual components by amplifying the influence of significant components and suppressing that of less influential components. The experimental results on benchmark datasets demonstrated that the proposed method achieved both high predictive accuracy and interpretability, making it suitable for practical use in synthetic pathway design for real-world applications.https://doi.org/10.1088/2632-2153/addfaachemical reaction yield predictiongraph neural networkinterpretable machine learningneural additive model
spellingShingle Youngchun Kwon
Yongsik Jung
Youn-Suk Choi
Seokho Kang
Interpretation of chemical reaction yields with graph neural additive network
Machine Learning: Science and Technology
chemical reaction yield prediction
graph neural network
interpretable machine learning
neural additive model
title Interpretation of chemical reaction yields with graph neural additive network
title_full Interpretation of chemical reaction yields with graph neural additive network
title_fullStr Interpretation of chemical reaction yields with graph neural additive network
title_full_unstemmed Interpretation of chemical reaction yields with graph neural additive network
title_short Interpretation of chemical reaction yields with graph neural additive network
title_sort interpretation of chemical reaction yields with graph neural additive network
topic chemical reaction yield prediction
graph neural network
interpretable machine learning
neural additive model
url https://doi.org/10.1088/2632-2153/addfaa
work_keys_str_mv AT youngchunkwon interpretationofchemicalreactionyieldswithgraphneuraladditivenetwork
AT yongsikjung interpretationofchemicalreactionyieldswithgraphneuraladditivenetwork
AT younsukchoi interpretationofchemicalreactionyieldswithgraphneuraladditivenetwork
AT seokhokang interpretationofchemicalreactionyieldswithgraphneuraladditivenetwork