{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T15:06:59Z","timestamp":1786460819884,"version":"build-2736575974"},"reference-count":53,"publisher":"American Chemical Society (ACS)","issue":"31","license":[{"start":{"date-parts":[[2026,7,30]],"date-time":"2026-07-30T00:00:00Z","timestamp":1785369600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100013890","name":"Ministerie van Landbouw, Natuur en Voedselkwaliteit","doi-asserted-by":"publisher","award":["KB-54-000-00"],"award-info":[{"award-number":["KB-54-000-00"]}],"id":[{"id":"10.13039\/501100013890","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2026,8,11]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Liquid chromatography coupled to high-resolution mass spectrometry (LC\u2013HRMS) is a widely used analytical technique for characterizing the chemical composition of organic samples. Due to its high sensitivity and ability to detect thousands of chemical features in a single run, untargeted LC\u2013HRMS experiments generate highly complex and data-rich datasets that typically require advanced computational methods, including machine learning, for meaningful interpretation. While traditional machine learning approaches have been applied to LC\u2013HRMS data, their performance remains limited for complex tasks. Deep learning has demonstrated improved performance, but both machine and deep learning are often constrained by the complexity and scarcity of labeled LC\u2013HRMS data. Foundation models present a promising new horizon for LC\u2013HRMS data analysis, given their ability to learn transferable representations from large-scale unlabeled data and adapt efficiently to downstream tasks with limited labeled samples. Recent studies have shown that foundation models can outperform conventional machine learning approaches in chemical annotation and molecular property prediction. We envision that foundation models for LC\u2013HRMS data will benefit from the expansion of curated sample repositories and spectral libraries, developing privacy-preserving training strategies, enabling simultaneous modeling of multiple LC\u2013HRMS data types, and improving model explainability.<\/jats:p>","DOI":"10.1021\/acs.analchem.6c01825","type":"journal-article","created":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T13:23:24Z","timestamp":1785849804000},"page":"22586-22594","source":"Crossref","is-referenced-by-count":0,"title":["Foundation\nModels for Liquid Chromatography\u2013High-Resolution\nMass Spectrometry: A New Era beyond Labeled Datasets"],"prefix":"10.1021","volume":"98","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8297-6212","authenticated-orcid":true,"given":"Andrea Junior","family":"Carnoli","sequence":"first","affiliation":[{"name":"Part of Wageningen University and Research , , ,","place":["Wageningen, The Netherlands, 6700 GE"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8300-6891","authenticated-orcid":true,"given":"Federico","family":"Padilla-Gonzalez","sequence":"additional","affiliation":[{"name":"Part of Wageningen University and Research , , ,","place":["Wageningen, The Netherlands, 6700 GE"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9123-7262","authenticated-orcid":true,"given":"Leonieke M.","family":"van den Bulk","sequence":"additional","affiliation":[{"name":"Part of Wageningen University and Research , , ,","place":["Wageningen, The Netherlands, 6700 GE"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Daan","family":"Korporaal","sequence":"additional","affiliation":[{"name":"Part of Wageningen University and Research , , ,","place":["Wageningen, The Netherlands, 6700 GE"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Martin","family":"Alewijn","sequence":"additional","affiliation":[{"name":"Part of Wageningen University and Research , , ,","place":["Wageningen, The Netherlands, 6700 GE"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3751-6065","authenticated-orcid":true,"given":"Marco H.","family":"Blokland","sequence":"additional","affiliation":[{"name":"Part of Wageningen University and Research , , ,","place":["Wageningen, The Netherlands, 6700 GE"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Bas H.M.","family":"van der Velden","sequence":"additional","affiliation":[{"name":"Part of Wageningen University and Research , , ,","place":["Wageningen, The Netherlands, 6700 GE"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"316","published-online":{"date-parts":[[2026,7,30]]},"reference":[{"issue":"36","key":"2026081109101923000_cit1","doi-asserted-by":"publisher","first-page":"12181","DOI":"10.1021\/acs.analchem.1c01309","article-title":"EVA: Evaluation of\nMetabolic Feature Fidelity Using a Deep Learning Model Trained with\nover 25000 Extracted Ion Chromatograms","volume":"93","author":"Guo","year":"2021","journal-title":"Anal.\nChem."},{"issue":"12","key":"2026081109101923000_cit2","doi-asserted-by":"publisher","first-page":"4930","DOI":"10.1021\/acs.analchem.1c02220","article-title":"Deep Learning-Assisted\nPeak Curation\nfor Large-Scale LC-MS Metabolomics","volume":"94","author":"Gloaguen","year":"2022","journal-title":"Anal. Chem."},{"issue":"2","key":"2026081109101923000_cit3","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pcbi.1008724","article-title":"Spec2Vec: Improved Mass Spectral\nSimilarity Scoring through Learning of Structural Relationships","volume":"17","author":"Huber","year":"2021","journal-title":"PLoS Comput. Biol."},{"issue":"1","key":"2026081109101923000_cit4","doi-asserted-by":"publisher","first-page":"84","DOI":"10.1186\/s13321-021-00558-4","article-title":"MS2DeepScore: A Novel\nDeep Learning Similarity Measure to Compare Tandem Mass Spectra","volume":"13","author":"Huber","year":"2021","journal-title":"J. Cheminform."},{"key":"2026081109101923000_cit5","article-title":"MS2Prop: A Machine\nLearning Model That Directly Generates de Novo Predictions of Drug-Likeness\nof Natural Products from Unannotated MS\/MS Spectra","author":"Voronov","year":"2022","journal-title":"bioRxiv"},{"issue":"1","key":"2026081109101923000_cit6","doi-asserted-by":"publisher","first-page":"1752","DOI":"10.1038\/s41467-023-37446-4","article-title":"MS2Query:\nreliable and scalable MS2 mass spectra-based analogue search","volume":"14","author":"de Jonge","year":"2023","journal-title":"Nat. Commun."},{"issue":"7","key":"2026081109101923000_cit7","doi-asserted-by":"publisher","first-page":"865","DOI":"10.1038\/s41592-022-01486-3","article-title":"MSNovelist: De Novo\nStructure Generation from Mass Spectra","volume":"19","author":"Stravs","year":"2022","journal-title":"Nat.\nMethods"},{"issue":"1","key":"2026081109101923000_cit8","doi-asserted-by":"publisher","first-page":"132","DOI":"10.1038\/s42004-023-00932-3","article-title":"An End-to-End\nDeep\nLearning Framework for Translating Mass Spectra to de-Novo Molecules","volume":"6","author":"Litsa","year":"2023","journal-title":"Commun. Chem."},{"key":"2026081109101923000_cit9","doi-asserted-by":"publisher","first-page":"538","DOI":"10.1038\/s41587-025-02677-x","article-title":"Self-Supervised Learning from Small-Molecule\nMass Spectrometry\nData","volume":"44","author":"Bittremieux","year":"2026","journal-title":"Nat. Biotechnol."},{"key":"2026081109101923000_cit10","doi-asserted-by":"publisher","first-page":"630","DOI":"10.1038\/s41587-025-02663-3","article-title":"Self-Supervised Learning\nof Molecular Representations\nfrom Millions of Tandem Mass Spectra Using DreaMS","volume":"44","author":"Bushuiev","year":"2026","journal-title":"Nat. Biotechnol."},{"key":"2026081109101923000_cit11","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-23190-2","volume-title":"Foundation Models for\nNatural Language Processing, Pre-trained Language Models Integrating\nMedia","author":"Paa\u00df","year":"2023"},{"issue":"7","key":"2026081109101923000_cit12","doi-asserted-by":"publisher","first-page":"1395","DOI":"10.1038\/s41592-025-02757-5","article-title":"Self-Supervised\nLearning of Molecular Representations","volume":"22","author":"Singh","year":"2025","journal-title":"Nat.\nMethods"},{"key":"2026081109101923000_cit13","doi-asserted-by":"publisher","first-page":"4111","DOI":"10.1007\/s00216-023-04776-7","article-title":"Critical Review on Data Processing\nAlgorithms in Non-Target\nScreening: Challenges and Opportunities to Improve Result Comparability","volume":"415","author":"Renner","year":"2023","journal-title":"Anal. Bioanal. Chem."},{"key":"2026081109101923000_cit14","doi-asserted-by":"publisher","first-page":"504","DOI":"10.1186\/1471-2105-9-504","article-title":"Highly Sensitive Feature\nDetection\nfor High Resolution LC\/MS","volume":"9","author":"Tautenhahn","year":"2008","journal-title":"BMC Bioinf."},{"issue":"17","key":"2026081109101923000_cit15","doi-asserted-by":"publisher","first-page":"8689","DOI":"10.1021\/acs.analchem.7b01069","article-title":"Detailed Investigation\nand Comparison of the XCMS and MZmine 2 Chromatogram Construction\nand Chromatographic Peak Detection Methods for Preprocessing Mass\nSpectrometry Metabolomics Data","volume":"89","author":"Myers","year":"2017","journal-title":"Anal. Chem."},{"issue":"1","key":"2026081109101923000_cit16","doi-asserted-by":"publisher","first-page":"588","DOI":"10.1021\/acs.analchem.9b04811","article-title":"Deep Learning for the Precise Peak\nDetection in High-Resolution LC-MS Data","volume":"92","author":"Melnikov","year":"2020","journal-title":"Anal.\nChem."},{"issue":"19","key":"2026081109101923000_cit17","doi-asserted-by":"publisher","first-page":"12407","DOI":"10.1021\/acs.analchem.9b02983","article-title":"Deep Neural\nNetworks for Classification of LC-MS Spectral Peaks","volume":"91","author":"Kantz","year":"2019","journal-title":"Anal. Chem."},{"key":"2026081109101923000_cit18","doi-asserted-by":"publisher","first-page":"255","DOI":"10.1016\/j.talanta.2019.05.053","article-title":"Using Deep Learning to Evaluate Peaks\nin Chromatographic Data","volume":"204","author":"Risum","year":"2019","journal-title":"Talanta"},{"issue":"1","key":"2026081109101923000_cit19","doi-asserted-by":"publisher","first-page":"151","DOI":"10.1109\/TCBB.2022.3141656","article-title":"An Interpretable\nDeep Learning Approach for Biomarker\nDetection in LC-MS Proteomics Data","volume":"20","author":"Iravani","year":"2023","journal-title":"IEEE\/ACM\nTrans. Comput. Biol. Bioinform."},{"issue":"1","key":"2026081109101923000_cit20","doi-asserted-by":"publisher","first-page":"7136","DOI":"10.1038\/s41467-024-51433-3","article-title":"An End-to-End Deep Learning Method\nfor Mass Spectrometry Data Analysis to Reveal Disease-Specific Metabolic\nProfiles","volume":"15","author":"Deng","year":"2024","journal-title":"Nat. Commun."},{"issue":"1","key":"2026081109101923000_cit21","doi-asserted-by":"publisher","first-page":"26705","DOI":"10.1038\/s41598-024-77494-4","article-title":"A Deep Learning Framework for Hepatocellular\nCarcinoma Diagnosis Using MS1 Data","volume":"14","author":"Xu","year":"2024","journal-title":"Sci. Rep."},{"key":"2026081109101923000_cit22","article-title":"MS-BART: Unified Modeling of Mass\nSpectra and Molecules\nfor Structure Elucidation","author":"Han","year":"2025","journal-title":"arXiv"},{"key":"2026081109101923000_cit23","article-title":"To Bin or Not\nto Bin: Alternative Representations of Mass Spectra","author":"de Jonge","year":"2025","journal-title":"arXiv"},{"key":"2026081109101923000_cit24","doi-asserted-by":"publisher","first-page":"bbag054","DOI":"10.1093\/bib\/bbag054","article-title":"ChemEmbed: A Deep Learning Framework\nfor Metabolite Identification Using Enhanced MS\/MS Data and Multidimensional\nMolecular Embeddings","volume":"27","author":"Faizan-Khan","year":"2026","journal-title":"Briefings Bioinf."},{"issue":"41","key":"2026081109101923000_cit25","doi-asserted-by":"publisher","first-page":"12580","DOI":"10.1073\/pnas.1509788112","article-title":"Searching\nMolecular Structure Databases with Tandem Mass Spectra Using CSI:\nFingerID","volume":"112","author":"Du\u0308hrkop","year":"2015","journal-title":"Proc. Natl. Acad. Sci. U. S. A."},{"issue":"7","key":"2026081109101923000_cit26","doi-asserted-by":"publisher","first-page":"754","DOI":"10.1038\/s42256-024-00856-0","article-title":"Molecular Set Representation Learning","volume":"6","author":"Boulougouri","year":"2024","journal-title":"Nat. Mach. Intell."},{"issue":"48","key":"2026081109101923000_cit27","doi-asserted-by":"publisher","first-page":"13738","DOI":"10.1073\/pnas.1608041113","article-title":"Topic Modeling for Untargeted Substructure Exploration\nin Metabolomics","volume":"113","author":"van der Hooft","year":"2016","journal-title":"Proc. Natl. Acad. Sci. U. S.\nA."},{"key":"2026081109101923000_cit28","article-title":"Large-Scale Discovery\nand Annotation of Hidden Substructure Patterns in Mass Spectrometry\nProfiles","author":"Torres-Ortega","year":"2025","journal-title":"bioRxiv"},{"key":"2026081109101923000_cit29","article-title":"Attention\nIs All You Need","author":"Vaswani","year":"2023","journal-title":"arXiv"},{"issue":"6","key":"2026081109101923000_cit30","doi-asserted-by":"publisher","first-page":"1097","DOI":"10.1007\/s11548-025-03355-8","article-title":"Foundation Model\nfor Assessing Cancer Tissue Margins\nwith Mass Spectrometry","volume":"20","author":"Farahmand","year":"2025","journal-title":"Int. J. Comput. Assist.\nRadiol. Surg."},{"issue":"42","key":"2026081109101923000_cit31","doi-asserted-by":"publisher","first-page":"16599","DOI":"10.1021\/acs.analchem.4c02426","article-title":"MSBERT: Embedding Tandem Mass Spectra\ninto Chemically Rational Space\nby Mask Learning and Contrastive Learning","volume":"96","author":"Zhang","year":"2024","journal-title":"Anal.\nChem."},{"key":"2026081109101923000_cit32","article-title":"LSM-MS2: A Foundation Model\nBridging\nSpectral Identification and Biological Interpretation","author":"Asher","year":"2025","journal-title":"arXiv"},{"key":"2026081109101923000_cit33","article-title":"LSM1-MS2: A Foundation\nModel for MS\/MS, Encompassing Chemical Property Predictions, Search\nand de Novo Generation","author":"Asher","year":"2024","journal-title":"ChemRxiv"},{"key":"2026081109101923000_cit34","unstructured":"() Healey, D.; Domingo-Ferna\u0301ndez, D.; Taylor, J.; Krettler, C.; Lightheart, R.; Park, T.; Kind, T.; Allen, A.; Colluru, V.PRISM: a Foundation Model For life\u2019s\nChemistryhttps:\/\/enveda.com\/prism-a-foundation-model-for-lifes-chemistry\/."},{"issue":"3","key":"2026081109101923000_cit35","doi-asserted-by":"publisher","first-page":"258","DOI":"10.1038\/s42256-022-00458-8","article-title":"Large Pre-Trained\nLanguage Models Contain Human-like Biases of What Is Right and Wrong\nto Do","volume":"4","author":"Schramowski","year":"2022","journal-title":"Nat. Mach. Intell."},{"key":"2026081109101923000_cit36","doi-asserted-by":"publisher","first-page":"88","DOI":"10.1186\/s13321-024-00878-1","article-title":"Reproducible MS\/MS Library Cleaning\nPipeline in Matchms","volume":"16","author":"de Jonge","year":"2024","journal-title":"J. Cheminform."},{"key":"2026081109101923000_cit37","article-title":"Towards Robust Foundation Models\nfor Digital Pathology","author":"Ko\u0308men","year":"2025","journal-title":"arXiv"},{"issue":"1","key":"2026081109101923000_cit38","doi-asserted-by":"publisher","first-page":"10780","DOI":"10.1038\/s41467-025-65823-8","article-title":"Benchmarking DNA Foundation Models\nfor Genomic and Genetic Tasks","volume":"16","author":"Feng","year":"2025","journal-title":"Nat. Commun."},{"key":"2026081109101923000_cit39","article-title":"Communication-Efficient\nLearning of Deep Networks from Decentralized Data","author":"McMahan","year":"2023","journal-title":"arXiv"},{"key":"2026081109101923000_cit40","doi-asserted-by":"publisher","first-page":"102545","DOI":"10.1016\/j.sbi.2023.102545","article-title":"Federated\nLearning for Molecular Discovery","volume":"79","author":"Hanser","year":"2023","journal-title":"Curr. Opin.\nStruct. Biol."},{"issue":"40","key":"2026081109101923000_cit41","doi-asserted-by":"publisher","first-page":"15840","DOI":"10.1021\/acs.analchem.4c02313","article-title":"FederEI: Federated\nLibrary Matching\nFramework for Electron Ionization Mass Spectrum Based Compound Identification","volume":"96","author":"Chen","year":"2024","journal-title":"Anal. Chem."},{"key":"2026081109101923000_cit42","doi-asserted-by":"publisher","first-page":"102238","DOI":"10.1016\/j.jafr.2025.102238","article-title":"Federated\nLearning in Food Research","volume":"23","author":"Fendor","year":"2025","journal-title":"J. Agric Food Res."},{"issue":"4","key":"2026081109101923000_cit43","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3510033","article-title":"FedBERT: When\nFederated Learning Meets Pre-Training","volume":"13","author":"Tian","year":"2022","journal-title":"ACM Trans.\nIntell. Syst. Technol."},{"key":"2026081109101923000_cit44","doi-asserted-by":"publisher","first-page":"343491","DOI":"10.1016\/j.aca.2024.343491","article-title":"Modular\nComparison of Untargeted Metabolomics Processing Steps","volume":"1336","author":"Aigensberger","year":"2025","journal-title":"Anal. Chim. Acta"},{"issue":"14","key":"2026081109101923000_cit45","doi-asserted-by":"publisher","first-page":"5894","DOI":"10.1021\/acs.analchem.2c04887","article-title":"Mechanistic\nUnderstanding of the Discrepancies between Common Peak\nPicking Algorithms in Liquid Chromatography-Mass Spectrometry-Based\nMetabolomics","volume":"95","author":"Guo","year":"2023","journal-title":"Anal. Chem."},{"issue":"27","key":"2026081109101923000_cit46","doi-asserted-by":"publisher","first-page":"6033","DOI":"10.1007\/s00216-024-05425-3","article-title":"A Stochastic Approach for Parameter Optimization of\nFeature Detection Algorithms for Non-Target Screening in Mass Spectrometry","volume":"417","author":"Sadia","year":"2025","journal-title":"Anal. Bioanal. Chem."},{"issue":"7","key":"2026081109101923000_cit47","doi-asserted-by":"publisher","first-page":"1447","DOI":"10.1038\/s41592-025-02718-y","article-title":"A Transformer Model for de Novo Sequencing\nof Data-Independent Acquisition Mass Spectrometry Data","volume":"22","author":"Sanders","year":"2025","journal-title":"Nat. Methods"},{"key":"2026081109101923000_cit48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01499","article-title":"Self-Supervised\nLearning from Images\nwith a Joint-Embedding Predictive Architecture","volume-title":"2023 IEEE\/CVF Conference On Computer Vision And Pattern Recognition\n(CVPR)","author":"Assran","year":"2023"},{"key":"2026081109101923000_cit49","doi-asserted-by":"publisher","first-page":"102470","DOI":"10.1016\/j.media.2022.102470","article-title":"Explainable\nArtificial Intelligence (XAI) in Deep Learning-Based Medical Image\nAnalysis","volume":"79","author":"van der Velden","year":"2022","journal-title":"Med. Image Anal."},{"key":"2026081109101923000_cit50","article-title":"A Unified Approach\nto Interpreting\nModel Predictions","author":"Lundberg","year":"2017","journal-title":"arXiv"},{"key":"2026081109101923000_cit51","article-title":"This Looks Like That: Deep Learning\nfor Interpretable\nImage Recognition","author":"Chen","year":"2019","journal-title":"arXiv"},{"issue":"12","key":"2026081109101923000_cit52","doi-asserted-by":"publisher","first-page":"3601","DOI":"10.1038\/s41591-024-03233-x","article-title":"A Foundation Model\nfor Clinician-Centered\nDrug Repurposing","volume":"30","author":"Huang","year":"2024","journal-title":"Nat. Med."},{"issue":"12","key":"2026081109101923000_cit53","doi-asserted-by":"publisher","first-page":"3422","DOI":"10.1038\/s41591-024-03333-8","article-title":"An Explainable Foundation\nModel for Drug Repurposing","volume":"30","author":"Bessadok","year":"2024","journal-title":"Nat. Med."}],"container-title":["Analytical Chemistry"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/pubs.acs.org\/doi\/pdf\/10.1021\/acs.analchem.6c01825","content-type":"application\/pdf","content-version":"vor","intended-application":"unspecified"},{"URL":"https:\/\/pubs.acs.org\/ancham\/article-pdf\/98\/31\/22586\/66343540\/acs.analchem.6c01825.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/pubs.acs.org\/ancham\/article-pdf\/98\/31\/22586\/66343540\/acs.analchem.6c01825.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,11]],"date-time":"2026-08-11T13:10:33Z","timestamp":1786453833000},"score":1,"resource":{"primary":{"URL":"https:\/\/pubs.acs.org\/ancham\/article\/98\/31\/22586\/5237437\/Foundation-Models-for-Liquid-Chromatography-High"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,30]]},"references-count":53,"journal-issue":{"issue":"31","published-online":{"date-parts":[[2026,7,30]]},"published-print":{"date-parts":[[2026,8,11]]}},"URL":"https:\/\/doi.org\/10.1021\/acs.analchem.6c01825","relation":{},"ISSN":["0003-2700","1520-6882"],"issn-type":[{"value":"0003-2700","type":"print"},{"value":"1520-6882","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,30]]}}}