\u30d6\u30fc\u30b9\u51fa\u5c55\u60c5\u5831<\/h2>\n\n\n\n
\u5c55\u793a\u671f\u9593<\/strong> : 2025\u5e748\u670817\u65e5\uff5e8\u670821\u65e5\uff08\u30a2\u30e1\u30ea\u30ab\u6771\u90e8\u6a19\u6e96\u6642\uff09 \u30bb\u30c3\u30b7\u30e7\u30f3:<\/strong> Poster Board #704 \u7c73\u6fa4 \u62d3\u5b5d<\/strong> \u30bb\u30c3\u30b7\u30e7\u30f3:<\/strong> Poster Board #928 In this study, we validated the accuracy of PFP for liquid phases. The solubility of organic liquids was evaluated using the Hildebrand solubility parameter derived from MD simulations. PFP agrees well with experimental data and predicts solubility more accurately than the GAFF force field. Furthermore, we validated the density of a water-ethanol mixture, with PFP showing good agreement with experimental data. Note that the underestimation for pure water is due to known errors in the PBE data set. It can be improved by training the model with more sophisticated functional. The results demonstrate that PFP is suitable for practical MD simulations and facilitates materials development.<\/p>\n\n\n\n \u540d\u5150\u8036 \u5f70\u6d0b<\/strong> \u30bb\u30c3\u30b7\u30e7\u30f3:<\/strong> Chemical Reaction Networks, Retrosynthesis, and Reaction Prediction \u30bb\u30c3\u30b7\u30e7\u30f3:<\/strong> Poster Board #922 <\/p>\n\n\n\n
\u958b\u50ac\u5834\u6240<\/strong> : Walter E. Washington Convention Center 801 Allen Y. Lew Place NW, Washington
\u30d6\u30fc\u30b9<\/strong> : 2045<\/p>\n\n\n\n\u30d7\u30ec\u30bc\u30f3\u30c6\u30fc\u30b7\u30e7\u30f3\u8a73\u7d30<\/h2>\n\n\n\n
\u30eb\u30fc\u30e0<\/strong>: Hall C (Walter E. Washington Convention Center)<\/p>\n\n\n\n\u767a\u8868\u8005 1: \u7c73\u6fa4 \u62d3\u5b5d (Hirotaka Yonezawa)<\/h3>\n\n\n\n
\u65e5\u6642:<\/strong> 2025\u5e748\u670820\u65e5 6:00 PM – 8:00 PM\uff08\u30a2\u30e1\u30ea\u30ab\u6771\u90e8\u6a19\u6e96\u6642\uff09
\u90e8\u9580:<\/strong> Division of Chemical Information <\/strong><\/strong>(CINF)
\u30bb\u30c3\u30b7\u30e7\u30f3\u30bf\u30a4\u30d7:<\/strong> \u30dd\u30b9\u30bf\u30fc\u30d7\u30ec\u30bc\u30f3\u30c6\u30fc\u30b7\u30e7\u30f3
\u30bf\u30a4\u30c8\u30eb:<\/strong> NMR chemical shift prediction with PFP descriptor: An application of universal machine learning interatomic potential
\u6982\u8981: <\/strong>Nuclear Magnetic Resonance (NMR) is a crucial technique for identifying the structure of organic molecules. Especially for complex molecules, it is essential to compare simulated spectra with measured data to assign the peaks and determine the structure accurately. This is why the research of chemical shift prediction has long been conducted.
Approaches to chemical shift prediction can be categorized into two types: ab initio techniques and data-driven methods. The latter can be further divided into three categories: additive increment-based, HOSE code, and machine learning.
Approaches have shown comparable good performance. However, these techniques share a common limitation: the representation of the surrounding environment beyond directly connected atoms.
In this study, we propose the approach to create NMR chemical shift prediction models with the descriptors extracted from machine learning interatomic potentials (MLIPs). This method is superior to previous methods in terms of representing the environment around atoms, especially geometrically distant atoms. This is because MLIPs already account for various intra\/inter- and short\/long-range atomic interactions and the descriptors can represent the complex interactions and local environments of atoms, providing a comprehensive basis for chemical shift prediction.<\/p>\n\n\n\n![\"Related](\"https:\/\/matlantis.com\/wp-content\/uploads\/2025\/08\/ACS-fall-2025_website_yonezawa-1024x459.png\")
<\/figure>\n\n\n\n
Matlantis\u682a\u5f0f\u4f1a\u793e \u30ab\u30b9\u30bf\u30de\u30fc\u30b5\u30af\u30bb\u30b9\u30a8\u30f3\u30b8\u30cb\u30a2\u3002
2018\u5e74\u306b\u6771\u4eac\u5927\u5b66\u3067\u535a\u58eb\uff08\u7406\u5b66\uff09\u3092\u53d6\u5f97\u5f8c\u3001\u30b3\u30f3\u30b5\u30eb\u30c6\u30a3\u30f3\u30b0\u30d5\u30a1\u30fc\u30e0\u306b\u3066\u30c7\u30fc\u30bf\u30b5\u30a4\u30a8\u30f3\u30b9\u306b\u3088\u308b\u696d\u52d9\u52b9\u7387\u5316\u306b\u5f93\u4e8b\u30022024\u5e74\u3088\u308aMatlantis\u306b\u5165\u793e\u3002\u5b9f\u9a13\u5316\u5b66\u8005\u3067\u3082\u8a08\u7b97\u306b\u3088\u308b\u7814\u7a76\u306e\u52b9\u7387\u5316\u304c\u3067\u304d\u308b\u4e16\u754c\u3092\u76ee\u6307\u3059\u3002\u5c02\u9580\u306f\u6709\u6a5f\u5408\u6210\u30fb\u8d85\u5206\u5b50\u30fb\u5149\u5316\u5b66\u3002<\/p>\n<\/div>\n\n\n\n![\"\"](\"https:\/\/matlantis.com\/wp-content\/uploads\/2025\/08\/portrait_hirotaka_yonezawa_up2.jpg\")
<\/figure>\n<\/div>\n<\/div>\n\n\n\n\u767a\u8868\u8005 2: \u540d\u5150\u8036 \u5f70\u6d0b (Akihiro Nagoya)<\/h3>\n\n\n\n
\u65e5\u6642:<\/strong> 2025\u5e748\u670819\u65e5 6:00 PM – 8:00 PM\uff08\u30a2\u30e1\u30ea\u30ab\u6771\u90e8\u6a19\u6e96\u6642\uff09
\u90e8\u9580:<\/strong> Division of Computers in Chemistry (COMP)
\u30bb\u30c3\u30b7\u30e7\u30f3\u30bf\u30a4\u30d7: <\/strong>\u30dd\u30b9\u30bf\u30fc\u30d7\u30ec\u30bc\u30f3\u30c6\u30fc\u30b7\u30e7\u30f3
\u30bf\u30a4\u30c8\u30eb: <\/strong>Application of a universal graph neural network potential for molecular dynamics of liquid phases
\u6982\u8981:<\/strong> Molecular dynamics (MD) simulations are essential tools for studying the structure and dynamics of liquid phase in chemistry, materials science and biophysics. Applications include solvation structure analysis, ion transport in electrolytes, phase equilibria, interfacial behaviour and diffusion processes. However, classical force fields often lack the accuracy required to capture complex interactions, such as chemical reactions at solid\/liquid interface, while ab initio MD is too computationally expensive.
Recent advances in machine learning potentials (MLPs) have opened new possibilities for liquid phase simulations. MLPs have been successfully applied to a number of liquid systems achieving DFT accuracy at much lower cost. MLPs allow MD simulations of complex phases that are inaccessible using quantum methods. On the other hand, the atomic configuration may be outside the training data set during long MD simulations, leading to error accumulation or inaccurate physical behaviour. The Preferred Potential (PFP), integrated into Matlantis\u2122, is a recently developed universal graph neural network (GNN) potential. PFP achieved a high level of accuracy and robustness due to its training on the huge DFT dataset. It offers superior applicability, transferability, and consistency across a broad range of systems without fine-tuning for specific systems. Thus, it is widely used for various applications such as catalysts, battery materials, metals and polymers.<\/p>\n\n\n\n![\"Related](\"https:\/\/matlantis.com\/wp-content\/uploads\/2025\/08\/ACS-fall-2025_website_nagoya-1024x477.png\")
<\/figure>\n\n\n\n
Matlantis\u682a\u5f0f\u4f1a\u793e \u30b7\u30cb\u30a2\u30a2\u30d7\u30ea\u30b1\u30fc\u30b7\u30e7\u30f3\u30b5\u30a4\u30a8\u30f3\u30c6\u30a3\u30b9\u30c8\u3002
\u5927\u962a\u5927\u5b66\u3092\u5352\u696d\u5f8c\u306b\u682a\u5f0f\u4f1a\u793e\u8c4a\u7530\u4e2d\u592e\u7814\u7a76\u6240\u306e\u7814\u7a76\u54e1\u3068\u3057\u3066\u52e4\u52d9\u3002\u592a\u967d\u96fb\u6c60\u6750\u6599\u3001\u71c3\u6599\u96fb\u6c60\u767d\u91d1\u89e6\u5a92\u3001\u4e8c\u6b21\u5143\u6750\u6599\u306e\u7b2c\u4e00\u539f\u7406\u8a08\u7b97\u3084\u9ad8\u5206\u5b50\u306e\u53e4\u5178MD\u8a08\u7b97\u3092\u4f7f\u7528\u3057\u305fMI\u306a\u3069\u306e\u7814\u7a76\u306b\u7d0415\u5e74\u9593\u5f93\u4e8b\u3057\u305f\u3002\u305d\u306e\u5f8c\u3001ENEOS\u682a\u5f0f\u4f1a\u793e\u306e\u4e2d\u592e\u6280\u8853\u7814\u7a76\u6240\u3092\u7d4c\u3066Matlantis\u306b\u5165\u793e\u3002\u73fe\u5728\u306f\u4e3b\u306b\u96fb\u6c60\u6750\u6599\u3084\u91d1\u5c5e\u6750\u6599\u306b\u95a2\u308f\u308b\u8a08\u7b97\u306b\u643a\u308f\u3063\u3066\u3044\u308b\u3002<\/p>\n<\/div>\n\n\n\n![\"\"](\"https:\/\/matlantis.com\/wp-content\/uploads\/2025\/08\/Akihiro_Nagoya_image-1.jpg\")
<\/figure>\n<\/div>\n<\/div>\n\n\n\n<\/h3>\n\n\n\n
Matlantis\u306b\u95a2\u3059\u308b\u793e\u5916\u306e\u8b1b\u6f14<\/h2>\n\n\n\n
2025\u5e748\u670818\u65e5<\/h3>\n\n\n\n
\u65e5\u6642:<\/strong> 2025\u5e748\u670818\u65e5 5:20 PM – 5:40 PM\uff08\u30a2\u30e1\u30ea\u30ab\u6771\u90e8\u6a19\u6e96\u6642\uff09
\u90e8\u9580:<\/strong> Division of Computers in Chemistry (COMP)
\u30eb\u30fc\u30e0:<\/strong> Hall E – Room 25 (Walter E. Washington Convention Center)
\u30bb\u30c3\u30b7\u30e7\u30f3\u30bf\u30a4\u30d7:<\/strong> \u53e3\u982d
\u30bf\u30a4\u30c8\u30eb:<\/strong> Automated generation of reaction pathways using neural networks
\u767a\u8868\u8005:<\/strong> Akihide Hayashi, So Takamoto, Ju Li, Hirotaka Akihide, Daisuka Okanohara
\u7d44\u7e54:<\/strong> Preferred Networks, inc. & Massachusetts Institute of Technology<\/p>\n\n\n\n2025\u5e748\u670819\u65e5<\/h3>\n\n\n\n
\u65e5\u6642:<\/strong> 2025\u5e748\u670819\u65e5 6:00 PM – 8:00 PM\uff08\u30a2\u30e1\u30ea\u30ab\u6771\u90e8\u6a19\u6e96\u6642\uff09
\u90e8\u9580:<\/strong> Division of Computers in Chemistry (COMP)
\u30eb\u30fc\u30e0:<\/strong> Hall C (Walter E. Washington Convention Center)
\u30bb\u30c3\u30b7\u30e7\u30f3\u30bf\u30a4\u30d7:<\/strong> \u30dd\u30b9\u30bf\u30fc\u30d7\u30ec\u30bc\u30f3\u30c6\u30fc\u30b7\u30e7\u30f3
\u30bf\u30a4\u30c8\u30eb:<\/strong> r2SCAN level universal neural network potential for molecules, crystals and surfaces
\u767a\u8868\u8005:<\/strong> Chikashi Shinagawa, So Takamoto, Daiki Shintani, Katsuhiko Nishimra, Kohei, Shinohara, Shigeru Iwase, Yuta Tsuboi, Ju Li
\u7d44\u7e54:<\/strong> Preferred Networks, inc. & Massachusetts Institute of Technology<\/p>\n\n\n\n