Scholarlink

<p>Graph Neural Networks (GNNs) are powerful tools for predicting chemical shifts in Nuclear Magnetic Resonance (NMR) spectroscopy. In this paper, we improve the state-of-the-art mean absolute error (MAE) on the Ilm-NMR-P31 dataset for the prediction of <sup>31</sup>P shifts from 11.4 ppm to 8.88 ppm by proposing a lightweight GNN which is based on the Metalayer-Framework. Furthermore, we analyze the performance of our model depending on the size of the training dataset and compare our model with different state-of-the-art models. Finally, we demonstrate how the GNN’s predictions can be interpreted and visualized with respect to the underlying molecular structures. Using GNNExplainer, we analyze the best- and worst-predicted molecules and perform feature ablations to assess the model’s reliance on specific input features. Furthermore, we validate the model’s physical plausibility by extracting learned substituent effects: the GNN autonomously rediscovers established empirical rules, accurately reproducing the <InlineEquation ID="IEq1"> <EquationSource Format="TEX">\(\beta\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>β</mi> </math></EquationSource> </InlineEquation>-deshielding and <InlineEquation ID="IEq2"> <EquationSource Format="TEX">\(\gamma\)</EquationSource> <EquationSource Format="MATHML"><math> <mi>γ</mi> </math></EquationSource> </InlineEquation>-shielding increments of neutral phosphines. Finally, we show that the SDGNN outperforms standard vacuum DFT calculations and HOSE code baselines.</p><p><b>Scientific Contribution:</b> The proposed lightweight GNN based on the Metalayer framework improves the state-of-the-art <sup>31</sup>P NMR shift prediction. By systematically varying the training-set size and benchmarking against multiple state-of-the-art models, we provide a standardized performance comparison that was previously lacking for <sup>31</sup>P NMR shift prediction. Using GNNExplainer and targeted feature ablations, we relate the model’s predictions to specific molecular substructures and input features. By quantitatively verifying that the model learns fundamental physical trends like substituent increments and identifying specific error sources, we provide a level of chemical interpretability and validation that goes beyond prior black-box approaches.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

A light-weight Graph Neural Network for the prediction of 31P Nuclear Magnetic Resonance signals

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月17日

查看全文
  • Dimitri Domnjuk,
  • Jana de Wiljes,
  • Robert Geitner

摘要

Graph Neural Networks (GNNs) are powerful tools for predicting chemical shifts in Nuclear Magnetic Resonance (NMR) spectroscopy. In this paper, we improve the state-of-the-art mean absolute error (MAE) on the Ilm-NMR-P31 dataset for the prediction of 31P shifts from 11.4 ppm to 8.88 ppm by proposing a lightweight GNN which is based on the Metalayer-Framework. Furthermore, we analyze the performance of our model depending on the size of the training dataset and compare our model with different state-of-the-art models. Finally, we demonstrate how the GNN’s predictions can be interpreted and visualized with respect to the underlying molecular structures. Using GNNExplainer, we analyze the best- and worst-predicted molecules and perform feature ablations to assess the model’s reliance on specific input features. Furthermore, we validate the model’s physical plausibility by extracting learned substituent effects: the GNN autonomously rediscovers established empirical rules, accurately reproducing the \(\beta\) β -deshielding and \(\gamma\) γ -shielding increments of neutral phosphines. Finally, we show that the SDGNN outperforms standard vacuum DFT calculations and HOSE code baselines.

Scientific Contribution: The proposed lightweight GNN based on the Metalayer framework improves the state-of-the-art 31P NMR shift prediction. By systematically varying the training-set size and benchmarking against multiple state-of-the-art models, we provide a standardized performance comparison that was previously lacking for 31P NMR shift prediction. Using GNNExplainer and targeted feature ablations, we relate the model’s predictions to specific molecular substructures and input features. By quantitatively verifying that the model learns fundamental physical trends like substituent increments and identifying specific error sources, we provide a level of chemical interpretability and validation that goes beyond prior black-box approaches.