[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-120826-en":3,"doc-seo-120826-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},120826,16904993612988,"Olivia Brown","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","Machine Learning Full NMR Chemical Shift Tensors of Silicon Oxides with Equivariant Graph Neural Networks - Research article summary","The nuclear magnetic resonance (NMR) chemical shift tensor encodes sensitive information about an atom’s electronic structure and its local environment. Many existing machine learning approaches focus only on predicting isotropic shifts, neglecting the tensorial content that carries richer structural details. This work employs an equivariant graph neural network to predict full 29Si chemical shift tensors in silicate materials, achieving strong accuracy for magnitude, anisotropy, and tensor orientation. Performance surpasses other ML models and analytic approaches, and the software is released as an open-source repository for reuse.","UC Berkeley  \nUC Berkeley Previously Published Works  \nTitle  \nMachine Learning Full NMR Chemical Shift Tensors of Silicon Oxides with Equivariant Graph Neural Networks  \nPermalink  \n[https://escholarship.org/uc/item/2h06b3w0](https://escholarship.org/uc/item/2h06b3w0)  \nJournal  \nThe Journal of Physical Chemistry A, 127(10)  \nISSN  \n1089-5639  \nAuthors  \nVenetos, Maxwell C  \nWen, Mingjian Persson, Kristin A  \nPublication Date  \n2023-03-16  \nDOI  \n10.1021/acs.jpca.2c07530  \nPeer reviewed  \n[eScholarship.org](eScholarship.org) Powered by the California Digital Library  \nUniversity of California  \n[pubs.acs.org/JPCA](pubs.acs.org/JPCA)  Article   \nMachine Learning Full NMR Chemical Shift Tensors of Silicon Oxides with Equivariant Graph Neural Networks  \nMaxwell C. Venetos, Mingjian Wen, and Kristin A. Persson*  \n Cite This: J. Phys. Chem. A 2023, 127, 2388−2398  \nRead Online  \n\n|  |  |  |  |  |  |\n| --- | --- | --- | --- | --- | --- |\n| ACCESS   | Metrics & More |  |  Article Recommendations |  | *sı Supporting Information |\n\nABSTRACT: The nuclear magnetic resonance (NMR) chemical shift tensor is a highly sensitive probe of the electronic structure of an atom and furthermore its local structure. Recently, machine learning has been applied to NMR in the prediction of isotropic chemical shifts from a structure. Current machine learning models, however, often ignore the full chemical shift tensor for the easier-topredict isotropic chemical shift, effectively ignoring a multitude of structural information available in the NMR chemical shift tensor. Here we use anequivariant graph neural network (GNN) to predict full 29Si chemical shift tensors in silicate materials. The equivariant GNN model predicts full tensors to a mean absolute error of 1.05 ppm and is able to accurately determine the magnitude, anisotropy, and tensor orientation in a diverse set of silicon oxide local structures. When compared with other models, the equivariant GNN model outperforms the state-of-the-art machine learning models by 53% . Theequivariant GNN model also outperforms historic analytical models by 57% for  \nisotropic chemical shift and 91% for anisotropy. The software is available as a simple-to-use open-source repository, allowing similar models to be created and trained with ease.  \n■ INTRODUCTION  \nMany useful properties of materials manifest from the precise structure of a given composition. Traditional structure determination techniques such as X-ray diffraction (XRD) are suitable for atoms of moderate to high atomic number; however, they can lead to ambiguous structure assignments for materials containing atoms with low atomic number. 1 In addition, XRD relies heavily on long-range order for correct measurement, but such long-range order is often lacking in many classes of materials, e.g., nanostructures, amorphous materials, and materials with a tetrahedral network, making structural characterization of such materials via XRD difficult. Such a class of materials are exemplified by silicates, which consist of a tetrahedral structure of silicons and oxygens (which have low atomic number and are difficult to observe via XRD).  \nSilicate materials are ubiquitous, from naturally occurring rocks and minerals like quartz2−5 and garnet6 to diverse manufacturing applications such as glasses,7,8 cements,9 and zeolite catalysts.10, 11 To discover new potential applications of silicates, accurate elucidation of their structures is a prerequisite.  \nNuclear magnetic resonance (NMR) spectroscopy has become a reliable tool for structural investigations in such materials. As a spectroscopy technique, NMR is highly sensitive to the electron density about an atom and relies on local structure rather than any long-range order. NMR measurements are typically combined with powder XRD  \nmeasurements and ab initio simulations to obtain refined crystal structures in a technique termed NMR crystallography. 12−22 These refinement procedures, however, often take an expens","cbCaijUjL8EyxvXR","https://ap.wps.com/l/cbCaijUjL8EyxvXR","pdf",2421249,1,12,"English","en",105,"# Abstract\n# Introduction\n## Limitations of conventional structure determination for silicates\n## Role of NMR crystallography and computational cost\n## Motivation for tensor-aware machine learning\n## Scope: predicting full 29Si chemical shift tensors","[{\"question\":\"Why are full NMR chemical shift tensors important beyond isotropic shifts?\",\"answer\":\"Isotropic chemical shifts capture only the scalar part of the tensor. Ignoring tensorial nature discards structural information contained in magnitude, anisotropy, and orientation.\"},{\"question\":\"What model is used to predict the 29Si chemical shift tensors?\",\"answer\":\"An equivariant graph neural network (GNN) is used to predict full 29Si chemical shift tensors from silicate local structures.\"},{\"question\":\"How does the proposed approach perform compared with other methods?\",\"answer\":\"The equivariant GNN predicts full tensors with a mean absolute error of 1.05 ppm, and it outperforms other machine learning models and historic analytical models for both isotropic chemical shift and anisotropy.\"}]","Machine Learning Full NMR Chemical Shift Tensors of Silicon Oxides with Equivariant Graph Neural Networks - Research article summary | PDF",1785732206,30,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"machine-learning-full-nmr-chemical-shift-tensors-of-silicon-oxides-with-equivariant-graph-neural-networks-research-article-summary","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/machine-learning-full-nmr-chemical-shift-tensors-of-silicon-oxides-with-equivariant-graph-neural-networks-research-article-summary/120826/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why are full NMR chemical shift tensors important beyond isotropic shifts?","Question",{"text":75,"@type":76},"Isotropic chemical shifts capture only the scalar part of the tensor. Ignoring tensorial nature discards structural information contained in magnitude, anisotropy, and orientation.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"What model is used to predict the 29Si chemical shift tensors?",{"text":80,"@type":76},"An equivariant graph neural network (GNN) is used to predict full 29Si chemical shift tensors from silicate local structures.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the proposed approach perform compared with other methods?",{"text":84,"@type":76},"The equivariant GNN predicts full tensors with a mean absolute error of 1.05 ppm, and it outperforms other machine learning models and historic analytical models for both isotropic chemical shift and anisotropy.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":29,"slug":121},"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]