{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T12:13:00Z","timestamp":1788437580576,"version":"build-2803163510"},"reference-count":44,"publisher":"American Chemical Society (ACS)","issue":"34","license":[{"start":{"date-parts":[[2026,8,21]],"date-time":"2026-08-21T00:00:00Z","timestamp":1787270400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001807","name":"Funda??o de Amparo ? i Pesquisa do Estado de S?o Paulo","doi-asserted-by":"publisher","award":["2014\/50279-4"],"award-info":[{"award-number":["2014\/50279-4"]}],"id":[{"id":"10.13039\/501100001807","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001807","name":"Funda??o de Amparo ? i Pesquisa do Estado de S?o Paulo","doi-asserted-by":"publisher","award":["2020\/15230-5"],"award-info":[{"award-number":["2020\/15230-5"]}],"id":[{"id":"10.13039\/501100001807","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001807","name":"Funda??o de Amparo ? i Pesquisa do Estado de S?o Paulo","doi-asserted-by":"publisher","award":["2021\/10501-3"],"award-info":[{"award-number":["2021\/10501-3"]}],"id":[{"id":"10.13039\/501100001807","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100002322","name":"Coordena??o de Aperfei?oamento de Pessoal de N?vel Superior","doi-asserted-by":"publisher","award":["NA"],"award-info":[{"award-number":["NA"]}],"id":[{"id":"10.13039\/501100002322","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100003593","name":"Conselho Nacional de Desenvolvimento Cient?fico e Tecnol?gico","doi-asserted-by":"publisher","award":["313430\/2023-3"],"award-info":[{"award-number":["313430\/2023-3"]}],"id":[{"id":"10.13039\/501100003593","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100005639","name":"Universidade de S?o Paulo","doi-asserted-by":"publisher","award":["NA"],"award-info":[{"award-number":["NA"]}],"id":[{"id":"10.13039\/501100005639","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100006487","name":"Ag?ncia Nacional do Petr?leo, G?s Natural e Biocombust?veis","doi-asserted-by":"publisher","award":["NA"],"award-info":[{"award-number":["NA"]}],"id":[{"id":"10.13039\/501100006487","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/na","name":"Petronas Petr?leo Brasil LTDA","doi-asserted-by":"publisher","award":["NA"],"award-info":[{"award-number":["NA"]}],"id":[{"id":"10.13039\/na","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/na","name":"Research Centre for Greenhouse Gas innovation","doi-asserted-by":"publisher","award":["NA"],"award-info":[{"award-number":["NA"]}],"id":[{"id":"10.13039\/na","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,9,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>We present Struct2ML, a scalable, low-latency volumetric inference pipeline that predicts three-dimensional electronic charge densities directly from atomic configurations. By leveraging a 3D U-Net architecture, our method provides a surrogate route to approximate charge-density fields without performing a full self-consistent Kohn\u2013Sham calculation for each queried structure, mapping multichannel structural descriptors with explicit periodic-boundary handling to volumetric charge fields. Unlike graph-based approaches that often require complex basis functions to reconstruct continuous fields, our grid-based representation maintains a computational complexity ofO(N), offering a natural interface for large-scale periodic and molecular systems optimized for real-time analysis. Key contributions include (i) scalable infrastructure, a high-throughput disk-to-GPU streaming dataloader optimized for terabyte-scale HDF5 archives, ensuring efficient training on massive data sets; (ii) robust training protocol, a reproducible workflow featuring automated voxel normalization, physics-informed multiobjective loss functions, and rigorous checkpointing; and (iii) multi-domain validation, a comprehensive evaluation suite providing both voxel-wise quantitative metrics and qualitative spatial analysis tailored for charge-density comparison. By substantially reducing the inference time to the millisecond scale (1\u201310 ms), this framework enables real-time charge-density exploration that reproduces key spatial and topological features of reference density functional theory (DFT) densities with data set-dependent accuracy, serving as a high-throughput utility complementary to equivariant models that primarily target static, maximal accuracy.<\/jats:p>","DOI":"10.1021\/acsomega.6c07986","type":"journal-article","created":{"date-parts":[[2026,8,22]],"date-time":"2026-08-22T13:02:11Z","timestamp":1787403731000},"page":"52123-52140","source":"Crossref","is-referenced-by-count":0,"title":["Struct2ML: End-to-End\nPrediction of Electronic Charge\nDensities Using 3D U-Net"],"prefix":"10.1021","volume":"11","author":[{"given":"Joa\u0303o Victor Vilela","family":"Cassiano","sequence":"first","affiliation":[{"name":"Universidade de Sa\u0303o Paulo, Instituto de Fi\u0301sica , Rua do Mata\u0303o, 1371 , , ,","place":["Sa\u0303o Paulo, SP, Brazil, 05508-090"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8599-2955","authenticated-orcid":true,"given":"Marcel","family":"Vale\u0301rio de Arruda","sequence":"additional","affiliation":[{"name":"Universidade de Sa\u0303o Paulo, Instituto de Fi\u0301sica , Rua do Mata\u0303o, 1371 , , ,","place":["Sa\u0303o Paulo, SP, Brazil, 05508-090"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8865-8082","authenticated-orcid":true,"given":"Alvaro David","family":"Torrez-Baptista","sequence":"additional","affiliation":[{"name":"Universidade de Sa\u0303o Paulo, Instituto de Fi\u0301sica , Rua do Mata\u0303o, 1371 , , ,","place":["Sa\u0303o Paulo, SP, Brazil, 05508-090"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8008-4907","authenticated-orcid":true,"given":"Caetano","family":"Rodrigues Miranda","sequence":"additional","affiliation":[{"name":"Universidade de Sa\u0303o Paulo, Instituto de Fi\u0301sica , Rua do Mata\u0303o, 1371 , , ,","place":["Sa\u0303o Paulo, SP, Brazil, 05508-090"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,8,21]]},"reference":[{"key":"2026090307164535100_cit1","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRev.136.B864","article-title":"Inhomogeneous Electron Gas","volume":"136","author":"Hohenberg","year":"1964","journal-title":"Phys. Rev."},{"key":"2026090307164535100_cit2","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRev.140.A1133","article-title":"Self-Consistent\nEquations Including Exchange and Correlation\nEffects","volume":"140","author":"Kohn","year":"1965","journal-title":"Phys. Rev."},{"key":"2026090307164535100_cit3","doi-asserted-by":"publisher","DOI":"10.1063\/1.4704546","article-title":"Perspective\non density functional theory","volume":"136","author":"Burke","year":"2012","journal-title":"J. Chem. Phys."},{"key":"2026090307164535100_cit4","doi-asserted-by":"publisher","DOI":"10.1103\/RevModPhys.87.897","article-title":"Density\nfunctional theory: Its origins, rise to prominence, and future","volume":"87","author":"Jones","year":"2015","journal-title":"Rev. Mod. Phys."},{"key":"2026090307164535100_cit5","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.54.11169","article-title":"Efficient iterative schemes for ab initio total-energy calculations using a plane-wave basis\nset","volume":"54","author":"Kresse","year":"1996","journal-title":"Phys. Rev. B"},{"key":"2026090307164535100_cit6","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.59.1758","article-title":"From ultrasoft pseudopotentials\nto the projector augmented-wave\nmethod","volume":"59","author":"Kresse","year":"1999","journal-title":"Phys. Rev. B"},{"key":"2026090307164535100_cit7","doi-asserted-by":"publisher","DOI":"10.1088\/0953-8984\/21\/39\/395502","article-title":"QUANTUM ESPRESSO: a\nmodular and open-source software project for quantum simulations of\nmaterials","volume":"21","author":"Giannozzi","year":"2009","journal-title":"J. Phys.: Condens. Matter"},{"key":"2026090307164535100_cit8","doi-asserted-by":"publisher","DOI":"10.1088\/0953-8984\/14\/11\/302","article-title":"The SIESTA\nmethod for ab initio order-N materials simulation","volume":"14","author":"Soler","year":"2002","journal-title":"J. Phys.: Condens. Matter"},{"key":"2026090307164535100_cit9","doi-asserted-by":"publisher","first-page":"2582","DOI":"10.1016\/j.cpc.2009.07.007","article-title":"ABINIT: First-principles\napproach to material\nand nanosystem properties","volume":"180","author":"Gonze","year":"2009","journal-title":"Comput. Phys. Commun."},{"key":"2026090307164535100_cit10","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.47.558","article-title":"Ab initio molecular dynamics for liquid\nmetals","volume":"47","author":"Kresse","year":"1993","journal-title":"Phys. Rev. B"},{"key":"2026090307164535100_cit11","doi-asserted-by":"publisher","first-page":"547","DOI":"10.1038\/s41586-018-0337-2","article-title":"Machine learning\nfor molecular and materials science","volume":"559","author":"Butler","year":"2018","journal-title":"Nature"},{"key":"2026090307164535100_cit12","doi-asserted-by":"publisher","DOI":"10.1038\/s41524-019-0221-0","article-title":"Recent advances\nand applications of machine learning in solid-state materials science","volume":"5","author":"Schmidt","year":"2019","journal-title":"npj Comput. Mater."},{"key":"2026090307164535100_cit13","doi-asserted-by":"publisher","first-page":"338","DOI":"10.1002\/inf2.12028","article-title":"Machine learning\nin materials science","volume":"1","author":"Wei","year":"2019","journal-title":"InfoMat"},{"key":"2026090307164535100_cit14","doi-asserted-by":"publisher","first-page":"5919","DOI":"10.1021\/acsomega.4c09861","article-title":"Search for Correlations\nBetween the\nResults of the Density Functional Theory and Hartree\u2013Fock Calculations\nUsing Neural Networks and Classical Machine Learning Algorithms","volume":"10","author":"Normatov","year":"2025","journal-title":"ACS Omega"},{"key":"2026090307164535100_cit15","doi-asserted-by":"publisher","first-page":"52010","DOI":"10.1021\/acsami.3c12050","article-title":"Electrochemical Sensor to Detect Antibiotics in Milk\nBased on Machine Learning Algorithms","volume":"15","author":"Aliev","year":"2023","journal-title":"ACS Appl.\nMater. Interfaces"},{"key":"2026090307164535100_cit16","doi-asserted-by":"publisher","first-page":"2017","DOI":"10.1021\/acs.jpclett.1c00007","article-title":"Programmable Soft-Matter\nElectronics","volume":"12","author":"Ivanov","year":"2021","journal-title":"J. Phys. Chem. Lett."},{"key":"2026090307164535100_cit17","doi-asserted-by":"publisher","DOI":"10.3390\/biomedicines11082251","article-title":"Machine\nLearning-Based Virtual Screening and Molecular Simulation Approaches\nIdentified Novel Potential Inhibitors for Cancer Therapy","volume":"11","author":"Shahab","year":"2023","journal-title":"Biomedicines"},{"key":"2026090307164535100_cit18","doi-asserted-by":"publisher","first-page":"459","DOI":"10.3390\/compounds3030034","article-title":"Computer\nModeling and Machine Learning in Chemistry and Materials Science:\nFrom Properties and Reactions of Small Organic and Inorganic Molecules\nto the Smart Design of Polymers and Composites","volume":"3","author":"Novikov","year":"2023","journal-title":"Compounds"},{"key":"2026090307164535100_cit19","doi-asserted-by":"publisher","DOI":"10.1063\/1.4812323","article-title":"Commentary: The Materials Project:\nA materials genome approach to accelerating materials innovation","volume":"1","author":"Jain","year":"2013","journal-title":"APL Mater."},{"key":"2026090307164535100_cit20","doi-asserted-by":"publisher","first-page":"314","DOI":"10.1016\/j.commatsci.2012.10.028","article-title":"Python Materials Genomics\n(pymatgen): A robust, open-source python\nlibrary for materials analysis","volume":"68","author":"Ong","year":"2013","journal-title":"Comput. Mater.\nSci."},{"key":"2026090307164535100_cit21","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-017-00839-3","article-title":"Bypassing the Kohn\u2013Sham equations\nwith machine learning","volume":"8","author":"Brockherde","year":"2017","journal-title":"Nat. Commun."},{"key":"2026090307164535100_cit22","doi-asserted-by":"publisher","DOI":"10.1016\/j.jcp.2021.110523","article-title":"Deep Density: circumventing\nthe Kohn\u2013Sham equations via symmetry preserving neural networks","volume":"443","author":"Zepeda-Nu\u0301n\u0303ez","year":"2021","journal-title":"J. Comput. Phys."},{"key":"2026090307164535100_cit23","doi-asserted-by":"publisher","first-page":"57","DOI":"10.1021\/acscentsci.8b00551","article-title":"Transferable Machine-Learning\nModel of the Electron\nDensity","volume":"5","author":"Grisafi","year":"2019","journal-title":"ACS Cent. Sci."},{"key":"2026090307164535100_cit24","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevB.100.184103","article-title":"Predicting charge density\ndistribution of materials using a local-environment-based graph convolutional\nnetwork","volume":"100","author":"Gong","year":"2019","journal-title":"Phys. Rev. B"},{"key":"2026090307164535100_cit25","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRevLett.125.206401","article-title":"Data-Driven\nWave Function,\nElectron Density Generation, and Atomization Energy Prediction and\nExtrapolation with Machine Learning","volume":"125","author":"Tsubaki","year":"2020","journal-title":"Phys. Rev.\nLett."},{"key":"2026090307164535100_cit26","doi-asserted-by":"publisher","DOI":"10.1038\/s41524-022-00863-y","article-title":"Equivariant graph neural\nnetworks for fast electron\ndensity estimation of molecules, liquids, and solids","volume":"8","author":"J\u00f8rgensen","year":"2022","journal-title":"npj Comput. Mater."},{"key":"2026090307164535100_cit27","unstructured":"Sinitskiy, A. V.; Pande, V. S.\n          Deep Neural\nNetwork Computes Electron Densities and Energies of a Large Set of\nOrganic Molecules Faster Than Density Functional Theory (DFT). 2018, arXiv:1809.02723. arXiv.org e-Print archive. https:\/\/arxiv.org\/abs\/1809.02723."},{"key":"2026090307164535100_cit28","unstructured":"J\u00f8rgensen, P. B.; Bhowmik, A.\n          DeepDFT:\nNeural Message Passing Network for Accurate Charge Density Prediction. 2020, arXiv:2011.03346. arXiv.org e-Print archive. https:\/\/arxiv.org\/abs\/2011.03346."},{"key":"2026090307164535100_cit29","doi-asserted-by":"publisher","DOI":"10.1038\/s41524-024-01343-1","article-title":"Higher-order\nequivariant neural networks for charge density prediction in materials","volume":"10","author":"Koker","year":"2024","journal-title":"npj Comput. Mater."},{"key":"2026090307164535100_cit30","doi-asserted-by":"crossref","unstructured":"Ronneberger, O.; Fischer, P.; Brox, T. In  U-Net: Convolutional\nNetworks for Biomedical Image Segmentation, Medical Image\nComputing and Computer-Assisted Intervention \u2013 MICCAI 2015; Springer International Publishing: Cham, 2015; pp 234\u2013241.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"2026090307164535100_cit31","doi-asserted-by":"crossref","unstructured":"C\u0327ic\u0327ek, O\u0308.; Abdulkadir, A.; Lienkamp, S. S.; Brox, T.; Ronneberger, O. In  3D U-Net:\nLearning Dense Volumetric Segmentation from Sparse Annotation, Medical Image Computing and Computer-Assisted Intervention \u2013\nMICCAI 2016; Springer International Publishing: Cham, 2016; pp 424\u2013432.","DOI":"10.1007\/978-3-319-46723-8_49"},{"key":"2026090307164535100_cit32","doi-asserted-by":"crossref","unstructured":"Milletari, F.; Navab, N.; Ahmadi, S.-A. In  V-Net: Fully Convolutional Neural\nNetworks for Volumetric Medical Image Segmentation, 2016\nFourth International Conference on 3D Vision (3DV); IEEE, 2016; pp 565\u2013571.","DOI":"10.1109\/3DV.2016.79"},{"key":"2026090307164535100_cit33","doi-asserted-by":"publisher","first-page":"297","DOI":"10.2307\/1932409","article-title":"Measures\nof the Amount of Ecologic Association Between Species","volume":"26","author":"Dice","year":"1945","journal-title":"Ecology"},{"key":"2026090307164535100_cit34","doi-asserted-by":"publisher","first-page":"600","DOI":"10.1109\/TIP.2003.819861","article-title":"Image quality\nassessment: From error visibility to structural similarity","volume":"13","author":"Wang","year":"2004","journal-title":"IEEE Trans. Image Process."},{"key":"2026090307164535100_cit35","doi-asserted-by":"crossref","unstructured":"Folk, M.; Heber, G.; Koziol, Q.; Pourmal, E.; Robinson, D. In  An Overview of the HDF5 Technology\nSuite and Its Applications, Proceedings of the EDBT\/ICDT\n2011 Workshop on Array Databases, 2011; pp 36\u201347.","DOI":"10.1145\/1966895.1966900"},{"key":"2026090307164535100_cit36","doi-asserted-by":"publisher","first-page":"261","DOI":"10.1038\/s41592-019-0686-2","article-title":"SciPy 1.0: fundamental\nalgorithms for scientific computing in Python","volume":"17","author":"Virtanen","year":"2020","journal-title":"Nat. Methods"},{"key":"2026090307164535100_cit37","doi-asserted-by":"publisher","first-page":"1145","DOI":"10.1109\/TC.2014.2349525","article-title":"Eliminating Periodic\nFlush Overhead of File I\/O with Non-Volatile Buffer Cache","volume":"65","author":"Lee","year":"2016","journal-title":"IEEE Trans. Comput."},{"key":"2026090307164535100_cit38","doi-asserted-by":"crossref","unstructured":"Lee, Y.-K.; Rhodes, W. T. In  Feature Detection\nand Enhancement by a Rotating Kernel Min-Max Transformation, Hybrid Image and Signal Processing II; SPIE, 1990; pp 154\u2013159.","DOI":"10.1117\/12.21310"},{"key":"2026090307164535100_cit39","doi-asserted-by":"crossref","unstructured":"Koishiyeva, D.; Alipbayev, K.; Kang, J. W.; Mukhamedgali, A.; Mukasheva, A. In  Comprehensive Evaluation\nof L2 Regularization in UNet-ResNet50 and UNet-VGG16 Models with Diverse\nImage Transformations, 2025 Artificial Intelligence, Computer,\nData Sciences and Applications (ACDSA); IEEE, 2025; pp 1\u20137.","DOI":"10.1109\/ACDSA65407.2025.11166386"},{"key":"2026090307164535100_cit40","doi-asserted-by":"crossref","unstructured":"Shah, S. A. A.; Nadeem, U.; Bennamoun, M.; Sohel, F.; Togneri, R. In  Efficient Image Set\nClassification\nUsing Linear Regression Based Image Reconstruction, Proceedings\nof the IEEE Conference on Computer Vision and Pattern Recognition\n(CVPR) Workshops; IEEE, 2017; pp 601\u2013610.","DOI":"10.1109\/CVPRW.2017.88"},{"key":"2026090307164535100_cit41","doi-asserted-by":"publisher","first-page":"98","DOI":"10.1109\/MSP.2008.930649","article-title":"Mean squared error: Love it or leave it? A new look\nat Signal Fidelity Measures","volume":"26","author":"Wang","year":"2009","journal-title":"IEEE Signal Process.\nMag."},{"key":"2026090307164535100_cit42","doi-asserted-by":"publisher","first-page":"3049","DOI":"10.1007\/s11042-017-5009-y","article-title":"A loss combination\nbased deep model for person re-identification","volume":"77","author":"Zhu","year":"2018","journal-title":"Multimed. Tools Appl."},{"key":"2026090307164535100_cit43","doi-asserted-by":"publisher","first-page":"792","DOI":"10.1109\/72.377990","article-title":"Improving model accuracy using optimal\nlinear combinations\nof trained neural networks","volume":"6","author":"Hashem","year":"1995","journal-title":"IEEE Trans. Neural\nNetw."},{"key":"2026090307164535100_cit44","unstructured":"Caljon, D.; Vercauteren, J.; Vos, S. D.; Verbeke, W.; Belle, J. V.\n          Using Dynamic Loss\nWeighting to Boost Improvements in Forecast Stability. 2025, arXiv:2409.18267. arXiv.org e-Print archive. https:\/\/arxiv.org\/abs\/2409.18267."}],"container-title":["ACS Omega"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acsomega.6c07986","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/acsodf\/article-pdf\/11\/34\/52123\/66984991\/acsomega.6c07986.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/acsodf\/article-pdf\/11\/34\/52123\/66984991\/acsomega.6c07986.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,3]],"date-time":"2026-09-03T11:16:53Z","timestamp":1788434213000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/acsodf\/article\/11\/34\/52123\/5285515\/Struct2ML-End-to-End-Prediction-of-Electronic"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,21]]},"references-count":44,"journal-issue":{"issue":"34","published-online":{"date-parts":[[2026,8,21]]},"published-print":{"date-parts":[[2026,9,1]]}},"URL":"https:\/\/doi.org\/10.1021\/acsomega.6c07986","relation":{},"ISSN":["2470-1343"],"issn-type":[{"value":"2470-1343","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,21]]}}}