Search

Menu
Search
in Ecology

Predicting missing links in food webs using stacked models and species traits

130 Views


Abstract

Networks are a powerful way to represent the complexity of large ecological systems. However, most ecological networks, such as food webs, contain only partial lists of species interactions. Computational methods for inferring missing links can facilitate field work and investigations of ecological processes. Here, we describe a stacked generalization approach to predict missing links in food webs that accounts for ecological assumptions including link direction. Tests of this method on synthetic food webs show that it can learn to optimally combine structural and trait-based predictions. On a global database of 290 food webs, the method often achieves near-perfect performance, performs better when it can exploit both species traits and network structure, and is principally driven by a subset of ecologically-interpretable predictors. Furthermore, we find that link predictability varies with ecosystem and network characteristics. These results show broad applicability of stacked generalization for predicting and understanding ecological interactions.

Similar content being viewed by others

A network simplification approach to ease topological studies about the food-web architecture

Article
Open access
17 August 2022

A species-level multi-trophic metaweb for Switzerland

Article
Open access
09 July 2025

Network rewiring conserves the topology of drought-impaired food webs

Article
Open access
24 November 2025

Data availability

The empirical food-web datasets used in this study are available in the GATEWAy database (GlobAl daTabasE of traits and food Web Architecture33,37, version 3, accessible at https://doi.org/10.25829/idiv.283-3-756). The data was pre-processed as described in Supplementary Note 2. The pre-processed data are available at https://doi.org/10.5281/zenodo.18026669. Supplementary Data 1 includes food web features across the datasets and summary statistics for food web features split by ecosystem type. Supplementary Data 2 includes details of Beta regression results.

Code availability

The code for this paper is available at https://doi.org/10.5281/zenodo.18026669 ref. 71. Data pre-processing and missing link prediction scripts were run with Python 3.6.3 on CentOS Linux 7 using High Performance Computing resources supported by BioFrontiers IT. Results visualization scripts were run on Windows 11 with Python 3.12.4. A configuration file is provided in the code repository for reproducing a Python environment with the necessary packages for running the scripts.

References

  1. Clauset, A., Moore, C. & Newman, M. E. J. Hierarchical structure and the prediction of missing links in networks. Nature 453, 98–101 (2008).

    Google Scholar 

  2. Ghasemian, A., Hosseinmardi, H., Galstyan, A., Airoldi, E. M. & Clauset, A. Stacking models for nearly optimal link prediction in complex networks. Proc. Natl. Acad. Sci. USA 117, 23393–23400 (2020).

    Google Scholar 

  3. Guimerà, R. & Sales-Pardo, M. Missing and spurious interactions and the reconstruction of complex networks. Proc. Natl. Acad. Sci. USA 106, 22073–22078 (2009).

    Google Scholar 

  4. Zhou, T. Progresses and challenges in link prediction. iScience 24, 103217 (2021).

    Google Scholar 

  5. Wolpert, D. H. Stacked generalization. Neural Netw. 5, 241–259 (1992).

    Google Scholar 

  6. Dunne, J. A., Brose, U., Williams, R. J. & Martinez, N. D. Modeling food-web dynamics: complexity-stability implications. In Aquatic food webs: an ecosystem approach, 117–129 (Oxford, 2005).

  7. McDonald-Madden, E. et al. Using food-web theory to conserve ecosystems. Nat. Commun. 7, 10245 (2016).

    Google Scholar 

  8. Kamenova, S. et al. Invasions Toolkit. In Advances in Ecological Research, vol. 56, 85–182 (Elsevier, 2017). https://linkinghub.elsevier.com/retrieve/pii/S0065250416300587.

  9. Van De Walle, R. et al. Arthropod food webs predicted from body size ratios are improved by incorporating prey defensive properties. J. Anim. Ecol. 92, 913–924 (2023).

    Google Scholar 

  10. Jordano, P. Sampling networks of ecological interactions. Funct. Ecol. 30, 1883–1893 (2016).

    Google Scholar 

  11. Terry, J. C. D. & Lewis, O. T. Finding missing links in interaction networks. Ecology 101, e03047 (2020).

    Google Scholar 

  12. Novak, M. et al. Predicting community responses to perturbations in the face of imperfect knowledge and network complexity. Ecology 92, 836–846 (2011).

    Google Scholar 

  13. Eklöf, A. et al. The dimensionality of ecological networks. Ecol. Lett. 16, 577–583 (2013).

    Google Scholar 

  14. Morales-Castilla, I., Matias, M. G., Gravel, D. & Araújo, M. B. Inferring biotic interactions from proxies. Trends Ecol. Evol. 30, 347–356 (2015).

    Google Scholar 

  15. Desjardins-Proulx, P., Laigle, I., Poisot, T. & Gravel, D. Ecological interactions and the Netflix problem. PeerJ 5, e3644 (2017).

    Google Scholar 

  16. Laigle, I. et al. Species traits as drivers of food web structure. Oikos 127, 316–326 (2018).

    Google Scholar 

  17. Pomeranz, J. P. F., Thompson, R. M., Poisot, T. & Harding, J. S. Inferring predator-prey interactions in food webs. Methods Ecol. Evol. 10, 356–367 (2019).

    Google Scholar 

  18. Gravel, D., Poisot, T., Albouy, C., Velez, L. & Mouillot, D. Inferring food web structure from predator-prey body size relationships. Methods Ecol. Evol. 4, 1083–1090 (2013).

    Google Scholar 

  19. Nagelkerke, L. A. J. & Rossberg, A. G. Trophic niche-space imaging, using resource and consumer traits. Theor. Ecol. 7, 423–434 (2014).

    Google Scholar 

  20. Wootton, K. L. et al. Layer-specific imprints of traits within a plant-herbivore-predator network – complementary insights from complementary methods. Ecography 2024, e07028 (2024).

    Google Scholar 

  21. Digel, C., Riede, J. O. & Brose, U. Body sizes, cumulative and allometric degree distributions across natural food webs. Oikos 120, 503–509 (2011).

    Google Scholar 

  22. Allesina, S. & Pascual, M. Food web models: a plea for groups. Ecol. Lett. 12, 652–662 (2009).

    Google Scholar 

  23. Williams, R. J. & Martinez, N. D. Simple rules yield complex food webs. Nature 404, 180–183 (2000).

    Google Scholar 

  24. Dalla Riva, G. V. & Stouffer, D. B. Exploring the evolutionary signature of food webs’ backbones using functional traits. Oikos 125, 446–456 (2016).

    Google Scholar 

  25. Rohr, R. P., Naisbit, R. E., Mazza, C. & Bersier, L.-F. Matching-centrality decomposition and the forecasting of new links in networks. Proc. R. Soc. B Biol. Sci. 283, 20152702 (2016).

    Google Scholar 

  26. Rohr, R., Scherer, H., Kehrli, P., Mazza, C. & Bersier, L. Modeling food webs: exploring unexplained structure using latent traits. Am. Naturalist 176, 170–177 (2010).

    Google Scholar 

  27. Williams, R. J., Anandanadesan, A. & Purves, D. The probabilistic niche model reveals the niche structure and role of body size in a complex food web. PLoS ONE 5, e12092 (2010).

    Google Scholar 

  28. Pearse, I. S. & Altermatt, F. Predicting novel trophic interactions in a non-native world. Ecol. Lett. 16, 1088–1094 (2013).

    Google Scholar 

  29. Stock, M., Poisot, T., Waegeman, W. & De Baets, B. Linear filtering reveals false negatives in species interaction data. Sci. Rep. 7, 45908 (2017).

    Google Scholar 

  30. Papadogeorgou, G., Bello, C., Ovaskainen, O. & Dunson, D. B. Covariate-informed latent interaction models: addressing geographic & taxonomic bias in predicting bird-plant interactions. J. Am. Stat. Assoc. 118, 2250–2261 (2023).

    Google Scholar 

  31. Poisot, T. Guidelines for the prediction of species interactions through binary classification. Methods Ecol. Evol. 14, 1333–1345 (2023).

    Google Scholar 

  32. Guimerà, R. One model to rule them all in network science? Proc. Natl. Acad. Sci. USA 117, 25195–25197 (2020).

    Google Scholar 

  33. Brose, U. GlobAL daTabasE of traits and food Web Architecture (GATEWAy) version 1.0 (2018). https://idata.idiv.de/ddm/Data/ShowData/283?version=3.

  34. Holland, P. W., Laskey, K. B. & Leinhardt, S. Stochastic blockmodels: first steps. Soc. Netw. 5, 109–137 (1983).

    Google Scholar 

  35. Penrose, M. Random geometric graphs (Oxford University Press, 2003). OCLC: 271204794.

  36. Ling, C. X., Huang, J. & Zhang, H. AUC: A Better Measure than Accuracy in Comparing Learning Algorithms. In Xiang, Y. & Chaib-draa, B. (eds.) Advances in Artificial Intelligence, vol. 2671, 329–341 (Springer Berlin Heidelberg, Berlin, Heidelberg, 2003). http://link.springer.com/10.1007/3-540-44886-1_25. Series Title: Lecture Notes in Computer Science.

  37. Brose, U. et al. Predator traits determine food-web architecture across ecosystems. Nat. Ecol. Evol. 3, 919–927 (2019).

    Google Scholar 

  38. Brimacombe, C., Bodner, K., Michalska-Smith, M., Poisot, T. & Fortin, M.-J. Shortcomings of reusing species interaction networks created by different sets of researchers. PLoS Biol. 21, e3002068 (2023).

    Google Scholar 

  39. Bohan, D. A., Caron-Lormier, G., Muggleton, S., Raybould, A. & Tamaddoni-Nezhad, A. Automated discovery of food webs from ecological data using logic-based machine learning. PLoS ONE 6, e29028 (2011).

    Google Scholar 

  40. Stouffer, D. B., Rezende, E. L. & Amaral, L. A. N. The role of body mass in diet contiguity and food-web structure: body mass and food-web structure. J. Anim. Ecol. 80, 632–639 (2011).

    Google Scholar 

  41. Huxham, M., Beaney, S. & Raffaelli, D. Do parasites reduce the chances of triangulation in a real food web? Oikos 76, 284 (1996).

    Google Scholar 

  42. Brose, U. et al. Consumer-resource body size relationships in natural food webs. Ecology 87, 2411–2417 (2006).

    Google Scholar 

  43. Pichler, M., Boreux, V., Klein, A., Schleuning, M. & Hartig, F. Machine learning algorithms to infer trait-matching and predict species interactions in ecological networks. Methods Ecol. Evol. 11, 281–293 (2020).

    Google Scholar 

  44. Caron, D. et al. Trait-matching models predict pairwise interactions across regions, not food web properties. Glob. Ecol. Biogeogr. 33, e13807 (2024).

    Google Scholar 

  45. Strydom, T. et al. A roadmap towards predicting species interaction networks (across space and time). Philos. Trans. R. Soc. B Biol. Sci. 376, 20210063 (2021).

    Google Scholar 

  46. Barner, A. K., Coblentz, K. E., Hacker, S. D. & Menge, B. A. Fundamental contradictions among observational and experimental estimates of non-trophic species interactions. Ecology 99, 557–566 (2018).

    Google Scholar 

  47. Blanchet, F. G., Cazelles, K. & Gravel, D. Co-occurrence is not evidence of ecological interactions. Ecol. Lett. 23, 1050–1063 (2020).

    Google Scholar 

  48. Li, J. et al. A size-constrained feeding-niche model distinguishes predation patterns between aquatic and terrestrial food webs. Ecol. Lett. 26, 76–86 (2023).

    Google Scholar 

  49. Cattin, M.-F., Bersier, L.-F., Banašek-Richter, C., Baltensperger, R. & Gabriel, J.-P. Phylogenetic constraints and adaptation explain food-web structure. Nature 427, 835–839 (2004).

    Google Scholar 

  50. Ives, A. & Godfray, H. Phylogenetic analysis of trophic associations. Am. Naturalist 168, E1–E14 (2006).

    Google Scholar 

  51. Petchey, O. L., Beckerman, A. P., Riede, J. O. & Warren, P. H. Size, foraging, and food web structure. Proc. Natl. Acad. Sci. USA 105, 4191–4196 (2008).

    Google Scholar 

  52. Grover, A. & Leskovec, J. node2vec: Scalable Feature Learning for Networks. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 855–864 (ACM, 2016). https://doi.org/10.1145/2939672.2939754.

  53. Strydom, T. et al. Graph embedding and transfer learning can help predict potential species interaction networks despite data limitations. Methods Ecol. Evol. 14, 2917–2930 (2023).

    Google Scholar 

  54. Lafferty, K. D. et al. Parasites in food webs: the ultimate missing links. Ecol. Lett. 11, 533–546 (2008).

    Google Scholar 

  55. Michalska-Smith, M. J. & Allesina, S. Telling ecological networks apart by their structure: a computational challenge. PLOS Comput. Biol. 15, e1007076 (2019).

    Google Scholar 

  56. Bar-Hen, A., Barbillon, P. & Donnet, S. Block models for generalized multipartite networks: applications in ecology and ethnobiology. Stat. Model. 22, 273–296 (2022).

    Google Scholar 

  57. Jacobs, A. Z., Dunne, J. A., Moore, C. & Clauset, A. Untangling the roles of parasites in food webs with generative network models (2015). Preprint, arxiv:1505.04741.

  58. Pringle, R. M. & Hutchinson, M. C. Resolving food-web structure. Annu. Rev. Ecol. Evol. Syst. 51, 55–80 (2020).

    Google Scholar 

  59. Tabouy, T., Barbillon, P. & Chiquet, J. Variational inference for stochastic block models from sampled data. J. Am. Stat. Assoc. 115, 455–466 (2020).

    Google Scholar 

  60. He, X. et al. Link prediction accuracy on real-world networks under non-uniform missing-edge patterns. PLoS ONE 19, e0306883 (2024).

    Google Scholar 

  61. Settles, B. From theories to queries: Active learning in practice. In Guyon, I., Cawley, G., Dror, G., Lemaire, V. & Statnikov, A. (eds.) Active Learning and Experimental Design workshop In conjunction with AISTATS 2010, vol. 16 of Proceedings of Machine Learning Research, 1–18 (PMLR, 2011). https://proceedings.mlr.press/v16/settles11a.html.

  62. Biton, B., Puzis, R. & Pilosof, S. Inductive link prediction facilitates the discovery of missing links and enables cross-community inference in ecological networks. Nature Ecology & Evolution 1–10 (Springer, 2025).

  63. Botella, C., Dray, S., Matias, C., Miele, V. & Thuiller, W. An appraisal of graph embeddings for comparing trophic network architectures. Methods Ecol. Evol. 13, 203–216 (2022).

    Google Scholar 

  64. He, X., Ghasemian, A., Lee, E., Clauset, A. & Mucha, P. J. Sequential stacking link prediction algorithms for temporal networks. Nat. Commun. 15, 1364 (2024).

    Google Scholar 

  65. Breiman, L. Random Forests. Mach. Learn. 45, 5–32 (2001).

    Google Scholar 

  66. Eklöf, A. & Ebenman, B. Species loss and secondary extinctions in simple and complex model communities. J. Anim. Ecol. 75, 239–246 (2006).

    Google Scholar 

  67. Liben-Nowell, D. & Kleinberg, J. The link-prediction problem for social networks. J. Am. Soc. Inf. Sci. Technol. 58, 1019–1031 (2007).

    Google Scholar 

  68. Cirtwill, A. R. & Wootton, K. L. Stable motifs delay species loss in simulated food webs. Oikos2022 (2022). https://onlinelibrary.wiley.com/doi/10.1111/oik.09436.

  69. Newman, M. E. J. & Clauset, A. Structure and inference in annotated networks. Nat. Commun. 7, 11863 (2016).

    Google Scholar 

  70. Cribari-Neto, F. & Zeileis, A. Beta regression in r. J. Stat. Softw. 34, 1–24 (2010).

    Google Scholar 

  71. Van Kleunen, L., Dee, L. E., Wootton, K. L., Massol, F. & Clauset, A. Predicting missing links in food webs using stacked models and species traits (2025). https://doi.org/10.5281/zenodo.18026669.

Download references

Acknowledgements

The authors thank the Brose lab for help with the empirical data, A. Ghasemian for helpful conversations about stacking models for link prediction, and B. Singh, D.B. Larremore and E. Bradley for helpful discussions and feedback. This work was supported in part by the National Science Foundation Division of Ocean Sciences (NSF OCE 2049360), the Eppley Foundation for Research, and the Chateaubriand Fellowship of the Office for Science & Technology of the Embassy of France in the United States. The stacking model code used here is adapted for use in directed, attributed, hierarchical networks from Ghasemian et al. (2020). The authors acknowledge the BioFrontiers Computing Core at the University of Colorado Boulder for providing High Performance Computing resources supported by BioFrontiers IT.

Author information

Authors and Affiliations

  1. Department of Computer Science, University of Colorado, Boulder, CO, USA

    Lucy B. Van Kleunen & Aaron Clauset

  2. Department of Public Health and Primary Care, KU Leuven Campus Kulak, Kortrijk, Belgium

    Lucy B. Van Kleunen

  3. Itec, imec research group at KU Leuven, Kortrijk, Belgium

    Lucy B. Van Kleunen

  4. Department of Ecology and Evolutionary Biology, University of Colorado, Boulder, CO, USA

    Laura E. Dee

  5. School of Biological Sciences, University of Canterbury, Christchurch, New Zealand

    Kate L. Wootton

  6. University of Lille, CNRS, Inserm, CHU Lille, Institut Pasteur de Lille, U1019 – UMR 9017 – CIIL – Center for Infection and Immunity of Lille, F-59000, Lille, France

    François Massol

  7. Sorbonne Université, Université Paris Cité, Université Paris Est Créteil, CNRS, INRAE, IRD, Institut d’Ecologie et des Sciences de l’Environnement de Paris (UMR7618), 75005, Paris, France

    François Massol

  8. BioFrontiers Institute, University of Colorado, Boulder, CO, USA

    Aaron Clauset

  9. Santa Fe Institute, Santa Fe, NM, USA

    Aaron Clauset

Authors

  1. Lucy B. Van Kleunen
    View author publications

    Search author on:PubMed Google Scholar

  2. Laura E. Dee
    View author publications

    Search author on:PubMed Google Scholar

  3. Kate L. Wootton
    View author publications

    Search author on:PubMed Google Scholar

  4. François Massol
    View author publications

    Search author on:PubMed Google Scholar

  5. Aaron Clauset
    View author publications

    Search author on:PubMed Google Scholar

Contributions

L.V., L.D., and A.C. conceptualized the study. L.D. and A.C. supervised the study. L.V. and A.C. designed the methodology and performed the mathematical analysis. L.V., L.D., and K.W. planned and validated the data pre-processing steps. L.V. wrote the code to pre-process the data, performed the computational analysis, and visualized results. L.D., K.W., F.M., and A.C. provided feedback to validate and improve the computational analysis and visualizations. L.V. and A.C. wrote an initial paper draft and all authors contributed to final paper writing, review, and editing.

Corresponding authors

Correspondence to
Lucy B. Van Kleunen or Aaron Clauset.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Communications Materials thanks Chencheng Cai, Virginia Dominguez-Garcia, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

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

Supplementary information

Supplementary Information

Description of Additional Supplementary Information

Supplementary Dataset 1

Supplementary Dataset 2

Reporting Summary

Transparent Peer Review file

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Van Kleunen, L.B., Dee, L.E., Wootton, K.L. et al. Predicting missing links in food webs using stacked models and species traits.
Nat Commun (2026). https://doi.org/10.1038/s41467-026-68769-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1038/s41467-026-68769-7

Source: Ecology - nature.com

Synergistic activity of sunlight protectants on some biocontrol agents as a new approach to control the cotton leaf worm, Spodoptera littoralis (Boisduval)

Seasonal bionomics of Anopheles mosquitoes in the cold-western Himalayan region of Kangra, Himachal Pradesh, India

This portal is not a newspaper as it is updated without periodicity. It cannot be considered an editorial product pursuant to law n. 62 of 7.03.2001. The author of the portal is not responsible for the content of comments to posts, the content of the linked sites. Some texts or images included in this portal are taken from the internet and, therefore, considered to be in the public domain; if their publication is violated, the copyright will be promptly communicated via e-mail. They will be immediately removed.

Back to Top
Close