Investigating Chemical Interpretability in Graph Neural Networks via Atom-wise Shapley Additive Explanations 


Vol. 69,  No. 4, pp. 165-176, Aug.  2025
10.5012/jkcs.2025.69.4.165


PDF JATS XML
  Abstract

The rapid advancement of machine learning (ML) has revolutionized molecular property predictions with achieving remarkable accuracy. However, their black-box nature limits interpretability, making it challenging for chemists to extract scientific insights and validate predictions against established chemical principles. To address this, Shapley Additive Explanations (SHAP) have been widely adopted, yet their application to graph neural networks (GNNs) remains challenging. Here, we develop a modified SHAP strategy to extract atom-wise contribution values from GNN predictions. We apply this approach to GNN models predicting fuel reactivity (cetane number) and Gibbs free energy of solvation. Our method provides chemically meaningful interpretations, aligning SHAP-derived descriptors with known chemical knowledge, including fuel's reactivity and solvation effects. The results demonstrate that atom-wise SHAP explanations offer valuable insights into molecular properties without requiring expensive quantum-mechanical calculations, enhancing the interpretability of ML-driven chemical predictions.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Cite this article

[IEEE Style]

H. Je, J. Kim, Y. Kim, "Investigating Chemical Interpretability in Graph Neural Networks via Atom-wise Shapley Additive Explanations," Journal of the Korean Chemical Society, vol. 69, no. 4, pp. 165-176, 2025. DOI: 10.5012/jkcs.2025.69.4.165.

[ACM Style]

Hyeonsu Je, JaeHun Kim, and Yeonjoon Kim. 2025. Investigating Chemical Interpretability in Graph Neural Networks via Atom-wise Shapley Additive Explanations. Journal of the Korean Chemical Society, 69, 4, (2025), 165-176. DOI: 10.5012/jkcs.2025.69.4.165.

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)

ISSN(Print) 1017-2548
ISSN(Online) 2234-8530