Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

  • Article
  • Published:

Pseudodata-based molecular structure generator to reveal unknown chemicals

Nature Machine Intelligence volume 7, pages 1879–1887 (2025)Cite this article

Subjects

Abstract

Translating mass spectra into chemical structures is a central challenge in exposomics, making it difficult to quickly track the millions of chemicals found in humans and the environment. Unlike metabolomics, key problems in developing models for chemicals with a larger molecular space include data scarcity, model complexity and proper query strategy. Here we present a molecular structure generator (MSGo) that can generate structures directly from mass spectra and discover unknown polyfluorinated chemicals in the exposome. Trained with only virtual spectra using a transformer neural network, MSGo correctly identified 48% of structures in a validation set and was better at discovering new polyfluorinated chemicals in wastewater samples reported in the literature than experts. Applying probability-oriented masking to the virtual spectra is key to MSGo’s performance. Rapid discovery of chemicals with limited experimental mass spectral data using automated tools such as MSGo is key to tackling the current unknown polyfluorinated chemical crisis.

This is a preview of subscription content, access via your institution

Access options

Buy this article

  • Purchase on SpringerLink
  • Instant access to the full article PDF.

USD 39.95

Prices may be subject to local taxes which are calculated during checkout

Fig. 1: Model architecture of MSGo.
Fig. 2: Validation of MSGo with experimental mass spectra from libraries.
Fig. 3: Validation of MSGo with mass spectra data from environmental samples.

Similar content being viewed by others

Data availability

All data needed to evaluate the conclusions in the paper are present in the paper or the Supplementary Information, and are available via GitHub at http://github.com/aaronma2020/MSGO and via Zenodo at https://doi.org/10.5281/zenodo.17182996 (ref. 46). Source data are provided with this paper.

Code availability

MSGo was developed using Python and is available for scientific research purposes via GitHub at http://github.com/aaronma2020/MSGO and Zenodo at https://doi.org/10.5281/zenodo.17182996 (ref. 46).

References

  1. Rappaport, S. M. & Smith, M. T. Environment and disease risks. Science 330, 460–461 (2010).

    Article  Google Scholar 

  2. Landrigan, P. J. et al. The Lancet Commission on pollution and health. Lancet 391, 462–512 (2018).

    Article  Google Scholar 

  3. Stein, S. E. & Scott, D. R. Optimization and testing of mass spectral library search algorithms for compound identification. J. Am. Soc. Mass. Spectrom. 5, 859–866 (1994).

    Article  Google Scholar 

  4. Scheubert, K. et al. Significance estimation for large scale metabolomics annotations by spectral matching. Nat. Commun. 8, 1494 (2017).

    Article  Google Scholar 

  5. Kind, T. et al. Identification of small molecules using accurate mass MS/MS search. Mass Spectrom. Rev. 37, 513–532 (2018).

    Article  Google Scholar 

  6. Dührkop, K., Shen, H., Meusel, M., Rousu, J. & Böcker, S. Searching molecular structure databases with tandem mass spectra using CSI:FingerID. Proc. Natl Acad. Sci. USA 112, 12580–12585 (2015).

    Article  Google Scholar 

  7. Ruttkies, C., Schymanski, E. L., Wolf, S., Hollender, J. & Neumann, S. MetFrag relaunched: incorporating strategies beyond in silico fragmentation. J. Cheminform. 8, 3 (2016).

    Article  Google Scholar 

  8. Tsugawa, H. et al. Hydrogen rearrangement rules: computational MS/MS fragmentation and structure elucidation using MS-FINDER software. Anal. Chem. 88, 7946–7958 (2016).

    Article  Google Scholar 

  9. Allen, F., Greiner, R. & Wishart, D. Competitive fragmentation modeling of ESI-MS/MS spectra for putative metabolite identification. Metabolomics 11, 98–110 (2015).

    Article  Google Scholar 

  10. Escher, B. I., Stapleton, H. M. & Schymanski, E. L. Tracking complex mixtures of chemicals in our changing environment. Science 367, 388–392 (2020).

    Article  Google Scholar 

  11. Vermeulen, R., Schymanski, E. L., Barabási, A.-L. & Miller, G. W. The exposome and health: where chemistry meets biology. Science 367, 392–396 (2020).

    Article  Google Scholar 

  12. Schymanski, E. L., Meinert, C., Meringer, M. & Brack, W. The use of MS classifiers and structure generation to assist in the identification of unknowns in effect-directed analysis. Anal. Chim. Acta 615, 136–147 (2008).

    Article  Google Scholar 

  13. Djoumbou-Feunang, Y. et al. BioTransformer: a comprehensive computational tool for small molecule metabolism prediction and metabolite identification. J. Cheminform. 11, 2 (2019).

    Article  Google Scholar 

  14. Reymond, J.-L. The chemical space project. Acc. Chem. Res. 48, 722–730 (2015).

    Article  Google Scholar 

  15. Moorthy, A. S., Wallace, W. E., Kearsley, A. J., Tchekhovskoi, D. V. & Stein, S. E. Combining fragment-ion and neutral-loss matching during mass spectral library searching: a new general purpose algorithm applicable to illicit drug identification. Anal. Chem. 89, 13261–13268 (2017).

    Article  Google Scholar 

  16. Xing, S. et al. Retrieving and utilizing hypothetical neutral losses from tandem mass spectra for spectral similarity analysis and unknown metabolite annotation. Anal. Chem. 92, 14476–14483 (2020).

    Article  Google Scholar 

  17. Aron, A. T. et al. Reproducible molecular networking of untargeted mass spectrometry data using GNPS. Nat. Protoc. 15, 1954–1991 (2020).

    Article  Google Scholar 

  18. Schmid, R. et al. Ion identity molecular networking for mass spectrometry-based metabolomics in the GNPS environment. Nat. Commun. 12, 3832 (2021).

    Article  Google Scholar 

  19. Tripathi, A. et al. Chemically informed analyses of metabolomics mass spectrometry data with Qemistree. Nat. Chem. Biol. 17, 146–151 (2021).

    Article  Google Scholar 

  20. Dührkop, K. et al. Systematic classification of unknown metabolites using high-resolution fragmentation mass spectra. Nat. Biotechnol. 39, 462–471 (2021).

    Article  Google Scholar 

  21. Vamathevan, J. et al. Applications of machine learning in drug discovery and development. Nat. Rev. Drug Discov. 18, 463–477 (2019).

    Article  Google Scholar 

  22. Colby, S. M., Nuñez, J. R., Hodas, N. O., Corley, C. D. & Renslow, R. R. Deep learning to generate in silico chemical property libraries and candidate molecules for small molecule identification in complex samples. Anal. Chem. 92, 1720–1729 (2019).

    Article  Google Scholar 

  23. Skinnider, M. A. et al. A deep generative model enables automated structure elucidation of novel psychoactive substances. Nat. Mach. Intell. 3, 973–984 (2021).

    Article  Google Scholar 

  24. Stravs, M. A., Dührkop, K., Böcker, S. & Zamboni, N. MSNovelist: de novo structure generation from mass spectra. Nat. Methods 19, 865–870 (2022).

    Article  Google Scholar 

  25. Shrivastava, A. D. et al. MassGenie: a transformer-based deep learning method for identifying small molecules from their mass spectra. Biomolecules 11, 1793 (2021).

    Article  Google Scholar 

  26. Litsa, E. E. et al. An end-to-end deep learning framework for translating mass spectra to de-novo molecules. Commun. Chem. 6, 132 (2023).

    Article  Google Scholar 

  27. Butler, T. et al. MS2Mol: a transformer model for illuminating dark chemical space from mass spectra. Preprint at https://doi.org/10.26434/chemrxiv-2023-vsmpx-v4.

  28. Madani, A. et al. Large language models generate functional protein sequences across diverse families. Nat. Biotechnol. 41, 1099–1106 (2023).

    Article  Google Scholar 

  29. Vaswani, A. et al. Attention is all you need. In Proc. 31st Conference on Neural Information Processing Systems (NIPS) (eds von Luxburg, U. et al.) 5999–6009 (NeruIPS, 2017).

  30. Blum, A. et al. The Madrid statement on poly-and perfluoroalkyl substances (PFASs). Environ. Health Perspect. 123, A107–A111 (2015).

    Article  Google Scholar 

  31. Evich, M. G. et al. Per- and polyfluoroalkyl substances in the environment. Science 375, eabg9065 (2022).

    Article  Google Scholar 

  32. Washington, J. W. et al. Nontargeted mass-spectral detection of chloroperfluoropolyether carboxylates in New Jersey soils. Science 368, 1103–1107 (2020).

    Article  Google Scholar 

  33. Djoumbou-Feunang, Y. et al. CFM-ID 3.0: significantly improved ESI-MS/MS prediction and compound identification. Metabolites 9, 72 (2019).

    Article  Google Scholar 

  34. Kong, F. et al. Denoising Search doubles the number of metabolite and exposome annotations in human plasma using an Orbitrap Astral mass spectrometer. Nat. Methods 22, 1008–1016 (2025).

    Article  Google Scholar 

  35. Li, X. & Fourches, D. SMILES pair encoding: a data-driven substructure tokenization algorithm for deep learning. J. Chem. Inf. Model. 61, 1560–1569 (2021).

    Article  Google Scholar 

  36. Dührkop, K. et al. SIRIUS 4: a rapid tool for turning tandem mass spectra into metabolite structure information. Nat. Methods 16, 299–302 (2019).

    Article  Google Scholar 

  37. Wang, Y. et al. Suspect and nontarget screening of per- and polyfluoroalkyl substances in wastewater from a fluorochemical manufacturing park. Environ. Sci. Technol. 52, 11007–11016 (2018).

    Article  Google Scholar 

  38. Fiehn Lab—CASMI 2022—Results (ucdavis.edu) (Univ. California Davis, 2022); https://fiehnlab.ucdavis.edu/casmi/casmi-2022-results

  39. Cai, Y., Zhou, Z. & Zhu, Z. J. Advanced analytical and informatic strategies for metabolite annotation in untargeted metabolomics. Trends Anal. Chem. 158, 116903 (2022).

    Article  Google Scholar 

  40. Lu, S., Gao, Z., He, D., Zhang, L. & Ke, G. Data-driven quantum chemical property prediction leveraging 3D conformations with Uni-Mol+. Nat. Commun. 15, 7104 (2024).

    Article  Google Scholar 

  41. Getzinger, G. J., Higgins, C. P. & Ferguson, P. L. Structure database and in silico spectral library for comprehensive suspect screening of per- and polyfluoroalkyl substances (PFASs) in environmental media by high-resolution mass spectrometry. Anal. Chem. 93, 2820–2827 (2021).

    Article  Google Scholar 

  42. Koelmel, J. P. et al. FluoroMatch 2.0—making automated and comprehensive non-targeted PFAS annotation a reality. Anal. Bioanal. Chem. 414, 1201–1215 (2022).

    Article  Google Scholar 

  43. Liu, Y., D’Agostino, L. A., Qu, G., Jiang, G. & Martin, J. W. High-resolution mass spectrometry (HRMS) methods for nontarget discovery and characterization of poly- and per-fluoroalkyl substances (PFASs) in environmental and human samples. Trends Anal. Chem. 121, 115420 (2019).

    Article  Google Scholar 

  44. Schymanski, E. L. et al. Identifying small molecules via high resolution mass spectrometry: communicating confidence. Environ. Sci. Technol. 48, 2097–2098 (2014).

    Article  Google Scholar 

  45. Reconciling Terminology of the Universe of Per- and Polyfluoroalkyl Substances: Recommendations and Practical Guidance (OECD, 2021); https://doi.org/10.1787/e458e796-en

  46. Ma, Z., Yu, N., Shao, Q., Bao, Q. & Wei, S. MSGO. Zenodo https://doi.org/10.5281/zenodo.17182996 (2025).

  47. Liu, M., Munoz, G., Duy, S. V., Sauvé, S. & Liu, J. Stability of nitrogen-containing polyfluoroalkyl substances in aerobic soils. Environ. Sci. Technol. 55, 4698–4708 (2021).

    Article  Google Scholar 

Download references

Acknowledgements

The MSGo project was supported by the National Key Research and Development Programme of China (grant no. 2024YFA0918900, X.W.), the National Natural Science Foundation of China (grant nos. 22525604, S.W., 22376092, S.W., U24A20512, S.W. and 22276090, N.Y.), the Fundamental Research Funds for the Central University (grant no. 021114380239, S.W.) and Anhui Provincial Key Research and Development Project (grant no. 2023t07020004, S.W.). We thank L. Wang, S. Yu and W. Jiang for their insights on polyfluorinated chemical synthesis; B. Zhang for constructive article feedback and Q. Bao for implementing benchmark models (Spec2Mol and MassGenie).

Author information

Author notes

  1. These authors contributed equally: Nanyang Yu, Zheng Ma, Qi Shao.

Authors and Affiliations

  1. State Key Laboratory of Water Pollution Control and Green Resource Recycling, School of Environment, Nanjing University, Nanjing, China

    Nanyang Yu, Qi Shao, Laihui Li, Xuebing Wang, Bingcai Pan, Hongxia Yu & Si Wei

  2. National Key Laboratory for Novel Software Technology, Nanjing University, Nanjing, China

    Zheng Ma

  3. Jiangsu Province Ecology and Environment Protection Key Laboratory of Chemical Safety and Health Risk, Nanjing University, Nanjing, China

    Qi Shao, Laihui Li, Xuebing Wang & Si Wei

Authors
  1. Nanyang Yu
    View author publications

    Search author on:PubMed Google Scholar

  2. Zheng Ma
    View author publications

    Search author on:PubMed Google Scholar

  3. Qi Shao
    View author publications

    Search author on:PubMed Google Scholar

  4. Laihui Li
    View author publications

    Search author on:PubMed Google Scholar

  5. Xuebing Wang
    View author publications

    Search author on:PubMed Google Scholar

  6. Bingcai Pan
    View author publications

    Search author on:PubMed Google Scholar

  7. Hongxia Yu
    View author publications

    Search author on:PubMed Google Scholar

  8. Si Wei
    View author publications

    Search author on:PubMed Google Scholar

Contributions

S.W., Z.M., Q.S. and N.Y. conceived the idea and designed the algorithm and software. Q.S. and Z.M. developed the MSGo program. Q.S., N.Y. and X.W. performed data collection and data processing. S.W., Z.M., Q.S. and N.Y. verified the performance of MSGo. S.W., Q.S. and N.Y. checked the structure annotation of MSGo. B.P. and X.W. suggested and designed synthesis routes. H.Y. and S.W. designed experiments to verify the annotated structure. N.Y. and Q.S. wrote the paper. S.W., N.Y., Z.M., Q.S., L.L., X.W., B.P. and H.Y. reviewed and edited the paper. S.W. supervised the project.

Corresponding author

Correspondence to Si Wei.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Machine Intelligence thanks Cheng Wang and the other, anonymous, reviewer(s) for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Supplementary Information

Supplementary Text 1, Figs. 1–16 and Table 1.

Reporting Summary

Supplementary Data

Supplementary Data 1. The molecular database for polyfluorinated chemicals. Supplementary Data 2. The experimental mass spectra dataset for polyfluorinated chemicals. Supplementary Data 3. The 90 reported structures in wastewater samples from the literature.

Source data

Source Data Fig. 2

Statistical source data.

Source Data Fig. 3

Statistical source data.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Yu, N., Ma, Z., Shao, Q. et al. Pseudodata-based molecular structure generator to reveal unknown chemicals. Nat Mach Intell 7, 1879–1887 (2025). https://doi.org/10.1038/s42256-025-01140-5

Download citation

  • Received:

  • Accepted:

  • Published:

  • Version of record:

  • Issue date:

  • DOI: https://doi.org/10.1038/s42256-025-01140-5

Search

Advanced search

Quick links

Nature Briefing: Translational Research

Sign up for the Nature Briefing: Translational Research newsletter — top stories in biotechnology, drug discovery and pharma.

Get what matters in translational research, free to your inbox weekly. Sign up for Nature Briefing: Translational Research