A KAN-based interpretable framework for prediction of global warming potential across chemical space
Accurate yet interpretable prediction of Global Warming Potential (GWP) is essential for the sustainable design of novel molecules, chemical processes and materials. This capability is valuable in the early-stage screening of compounds with potential relevance to carbon management and emerging CCUS...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-09-01
|
| Series: | Carbon Capture Science & Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2772656825001174 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849239532969394176 |
|---|---|
| author | Jaewook Lee Xinyang Sun Ethan Errington Calum Drysdale Miao Guo |
| author_facet | Jaewook Lee Xinyang Sun Ethan Errington Calum Drysdale Miao Guo |
| author_sort | Jaewook Lee |
| collection | DOAJ |
| description | Accurate yet interpretable prediction of Global Warming Potential (GWP) is essential for the sustainable design of novel molecules, chemical processes and materials. This capability is valuable in the early-stage screening of compounds with potential relevance to carbon management and emerging CCUS applications. However, conventional models often face a trade-off between predictive accuracy and interpretability. In this study, we propose an AI-based GWP prediction framework that integrates both molecular and process-level features to improve accuracy while employing white-box modeling techniques to enhance interpretability. First, by incorporating molecular descriptors (MACCS keys, Mordred descriptors) and process-level information (process title, description, location), the Deep Neural Network (DNN) model achieved an R² of 86 % on the test data, representing a 25 % improvement over the most comparable benchmark reported in prior studies. XAI analysis further highlights the crucial role of process-related features, particularly process title embeddings, in enhancing model predictions. Second, to address the need for model transparency, we employed a Kolmogorov–Arnold Network (KAN) model to develop a symbolic, white-box GWP prediction model. While achieving a lower R² of 64 %, this model provides explicit mathematical representations of GWP relationships, enabling interpretable decision-making in sustainable chemical and process design. Our findings demonstrate that integrating molecular and process-level features improves both predictive accuracy and interpretability in GWP modelling. The resulting framework can support early-stage environmental assessment of novel compounds, offering a useful tool to inform the sustainable design of chemicals, including those with potential applications in CCUS. |
| format | Article |
| id | doaj-art-ac1b87ef593f41e1aaa383e541564932 |
| institution | Kabale University |
| issn | 2772-6568 |
| language | English |
| publishDate | 2025-09-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Carbon Capture Science & Technology |
| spelling | doaj-art-ac1b87ef593f41e1aaa383e5415649322025-08-20T04:00:55ZengElsevierCarbon Capture Science & Technology2772-65682025-09-011610047810.1016/j.ccst.2025.100478A KAN-based interpretable framework for prediction of global warming potential across chemical spaceJaewook Lee0Xinyang Sun1Ethan Errington2Calum Drysdale3Miao Guo4Department of Engineering, King’s College London, London WC2R 2LS, United KingdomDepartment of Engineering, King’s College London, London WC2R 2LS, United KingdomDepartment of Engineering, King’s College London, London WC2R 2LS, United KingdomDepartment of Engineering, King’s College London, London WC2R 2LS, United KingdomCorresponding author.; Department of Engineering, King’s College London, London WC2R 2LS, United KingdomAccurate yet interpretable prediction of Global Warming Potential (GWP) is essential for the sustainable design of novel molecules, chemical processes and materials. This capability is valuable in the early-stage screening of compounds with potential relevance to carbon management and emerging CCUS applications. However, conventional models often face a trade-off between predictive accuracy and interpretability. In this study, we propose an AI-based GWP prediction framework that integrates both molecular and process-level features to improve accuracy while employing white-box modeling techniques to enhance interpretability. First, by incorporating molecular descriptors (MACCS keys, Mordred descriptors) and process-level information (process title, description, location), the Deep Neural Network (DNN) model achieved an R² of 86 % on the test data, representing a 25 % improvement over the most comparable benchmark reported in prior studies. XAI analysis further highlights the crucial role of process-related features, particularly process title embeddings, in enhancing model predictions. Second, to address the need for model transparency, we employed a Kolmogorov–Arnold Network (KAN) model to develop a symbolic, white-box GWP prediction model. While achieving a lower R² of 64 %, this model provides explicit mathematical representations of GWP relationships, enabling interpretable decision-making in sustainable chemical and process design. Our findings demonstrate that integrating molecular and process-level features improves both predictive accuracy and interpretability in GWP modelling. The resulting framework can support early-stage environmental assessment of novel compounds, offering a useful tool to inform the sustainable design of chemicals, including those with potential applications in CCUS.http://www.sciencedirect.com/science/article/pii/S2772656825001174Life cycle assessment (LCA)Global warming potential (GWP)Explainable artificial intelligence (XAI)Kolmogorov-Arnold network (KAN) |
| spellingShingle | Jaewook Lee Xinyang Sun Ethan Errington Calum Drysdale Miao Guo A KAN-based interpretable framework for prediction of global warming potential across chemical space Carbon Capture Science & Technology Life cycle assessment (LCA) Global warming potential (GWP) Explainable artificial intelligence (XAI) Kolmogorov-Arnold network (KAN) |
| title | A KAN-based interpretable framework for prediction of global warming potential across chemical space |
| title_full | A KAN-based interpretable framework for prediction of global warming potential across chemical space |
| title_fullStr | A KAN-based interpretable framework for prediction of global warming potential across chemical space |
| title_full_unstemmed | A KAN-based interpretable framework for prediction of global warming potential across chemical space |
| title_short | A KAN-based interpretable framework for prediction of global warming potential across chemical space |
| title_sort | kan based interpretable framework for prediction of global warming potential across chemical space |
| topic | Life cycle assessment (LCA) Global warming potential (GWP) Explainable artificial intelligence (XAI) Kolmogorov-Arnold network (KAN) |
| url | http://www.sciencedirect.com/science/article/pii/S2772656825001174 |
| work_keys_str_mv | AT jaewooklee akanbasedinterpretableframeworkforpredictionofglobalwarmingpotentialacrosschemicalspace AT xinyangsun akanbasedinterpretableframeworkforpredictionofglobalwarmingpotentialacrosschemicalspace AT ethanerrington akanbasedinterpretableframeworkforpredictionofglobalwarmingpotentialacrosschemicalspace AT calumdrysdale akanbasedinterpretableframeworkforpredictionofglobalwarmingpotentialacrosschemicalspace AT miaoguo akanbasedinterpretableframeworkforpredictionofglobalwarmingpotentialacrosschemicalspace AT jaewooklee kanbasedinterpretableframeworkforpredictionofglobalwarmingpotentialacrosschemicalspace AT xinyangsun kanbasedinterpretableframeworkforpredictionofglobalwarmingpotentialacrosschemicalspace AT ethanerrington kanbasedinterpretableframeworkforpredictionofglobalwarmingpotentialacrosschemicalspace AT calumdrysdale kanbasedinterpretableframeworkforpredictionofglobalwarmingpotentialacrosschemicalspace AT miaoguo kanbasedinterpretableframeworkforpredictionofglobalwarmingpotentialacrosschemicalspace |