MELLODDY: Cross-pharma Federated Learning at Unprecedented Scale Unlocks Benefits in QSAR without Compromising Proprietary Information
Open accessPublished version of a preprint
The DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index.
Abstract
Federated multipartner machine learning has been touted as an appealing and efficient method to increase the effective training data volume and thereby the predictivity of models, particularly when the generation of training data is resource-intensive. In the landmark MELLODDY project, indeed, each of ten pharmaceutical companies realized aggregated improvements on its own classification or regression models through federated learning. To this end, they leveraged a novel implementation extending multitask learning across partners, on a platform audited for privacy and security. The experiments involved an unprecedented cross-pharma data set of 2.6+ billion confidential experimental activity data points, documenting 21+ million physical small molecules and 40+ thousand assays in on-target and secondary pharmacodynamics and pharmacokinetics. Appropriate complementary metrics were developed to evaluate the predictive performance in the federated setting. In addition to predictive performance increases in labeled space, the results point toward an extended applicability domain in federated learning. Increases in collective training data volume, including by means of auxiliary data resulting from single concentration high-throughput and imaging assays, continued to boost predictive performance, albeit with a saturating return. Markedly higher improvements were observed for the pharmacokinetics and safety panel assay-based task subsets.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Chapter 12: What does the Anthropocene hold for citizenship?
(2019) · edoc (University of Basel) · doi:10.1007/978-3 · Cited by 11,533
The relationship between Precision-Recall and ROC curves
Davis, J., & Goadrich, M. (2006) · Proceedings of the 23rd international conference on Machine learning - ICML '06 · doi:10.1145/1143844.1143874 · Cited by 6,559
Federated Machine Learning: Concept and Applications
Yang, Q., Liu, Y., Chen, T., & Tong, Y. (2019) · ACM Transactions on Intelligent Systems and Technology · doi:10.1145/3298981 · Cited by 6,322
Chicco, D., Warrens, M. J., & Jurman, G. (2021) · PeerJ Computer Science · doi:10.7717/peerj-cs.623 · Cited by 5,380
Saito, T., & Rehmsmeier, M. (2015) · PLOS ONE · doi:10.1371/journal.pone.0118432 · Cited by 5,309
p-σ-π Analysis. A Method for the Correlation of Biological Activity and Chemical Structure
Hansch, C., & Fujita, T. (1964) · Journal of the American Chemical Society · doi:10.1021/ja01062a035 · Cited by 2,313
Analyzing Learned Molecular Representations for Property Prediction
Yang, K., Swanson, K., Jin, W., Coley, C., Eiden, P., Gao, H., Guzman-Perez, A., Hopper, T., Kelley, B., Mathea, M., et al. (2019) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.9b00237 · Cited by 1,923
Deep Neural Nets as a Method for Quantitative Structure–Activity Relationships
Ma, J., Sheridan, R. P., Liaw, A., Dahl, G. E., & Svetnik, V. (2015) · Journal of Chemical Information and Modeling · doi:10.1021/ci500747n · Cited by 1,159
HANSCH, C., MALONEY, P. P., FUJITA, T., & MUIR, R. M. (1962) · Nature · doi:10.1038/194178b0 · Cited by 1,056
Muratov, E. N., Bajorath, J., Sheridan, R. P., Tetko, I. V., Filimonov, D., Poroikov, V., Oprea, T. I., Baskin, I. I., Varnek, A., Roitberg, A., et al. (2020) · Chemical Society Reviews · doi:10.1039/d0cs00098a · Cited by 913
Reducing safety-related drug attrition: the use of in vitro pharmacological profiling
Bowes, J., Brown, A. J., Hamon, J., Jarolimek, W., Sridhar, A., Waldron, G., & Whitebread, S. (2012) · Nature Reviews Drug Discovery · doi:10.1038/nrd3845 · Cited by 748 · cited 2×
“The next sections will present a detailed breakdown of the federated performance gains, starting with an analysis of the classification models in terms of predictive performance and applicability domain metrics, followed by a comparison to the regression model performance.”
BackgroundMethodLarge-scale comparison of machine learning methods for drug target prediction on ChEMBL
Mayr, A., Klambauer, G., Unterthiner, T., Steijaert, M., Wegner, J. K., Ceulemans, H., Clevert, D.-A., & Hochreiter, S. (2018) · Chemical Science · doi:10.1039/c8sc00148k · Cited by 584
Self-citationI Have a DREAM! (DiffeRentially privatE smArt Metering)
Ács, G., & Castelluccia, C. (2011) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-24178-9_9 · Cited by 388 · cited 2×
“24,25 On the platform, the weights of the common trunk could be trained in a federated way by applying secure aggregation 26 of the individual gradients from each minibatch of the contributing partners.”
BackgroundMethodSelf-citationTime-Split Cross-Validation as a Method for Estimating the Goodness of Prospective Prediction.
Sheridan, R. P. (2013) · Journal of Chemical Information and Modeling · doi:10.1021/ci400084k · Cited by 376
Lenselink, E. B., ten Dijke, N., Bongers, B., Papadatos, G., van Vlijmen, H. W. T., Kowalczyk, W., IJzerman, A. P., & van Westen, G. J. P. (2017) · Journal of Cheminformatics · doi:10.1186/s13321-017-0232-0 · Cited by 334
Algorithmic Learning in a Random World
Vovk, V., Gammerman, A., & Shafer, G. (2022) · doi:10.1007/978-3-031-06649-8 · Cited by 296
Is Multitask Deep Learning Practical for Pharma?
Ramsundar, B., Liu, B., Wu, Z., Verras, A., Tudor, M., Sheridan, R. P., & Pande, V. (2017) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.7b00146 · Cited by 263
Demystifying Multitask Deep Neural Networks for Quantitative Structure–Activity Relationships
Xu, Y., Ma, J., Liaw, A., Sheridan, R. P., & Svetnik, V. (2017) · Journal of Chemical Information and Modeling · doi:10.1021/acs.jcim.7b00087 · Cited by 238
Bayer’s in silico ADMET platform: a journey of machine learning over the past two decades
Göller, A. H., Kuhnke, L., Montanari, F., Bonin, A., Schneckener, S., ter Laak, A., Wichard, J., Lobell, M., & Hillisch, A. (2020) · Drug Discovery Today · doi:10.1016/j.drudis.2020.07.001 · Cited by 215
Self-citation
Research works printed but not linked: 33
- 10.1016/S0223-5234(98)80023-8 · doi:10.1016/s0223-5234(98)80023-8
- 10.1016/j.bmc.2006.05.018 · DOI
- Unterthiner T., 2014, Adv. Neural Inf. Process. Syst., 27, 1
- Dahl, G. E.; Jaitly, N.; Salakhutdinov, R. Multi-Task Neural Networks for QSAR Predictions. arXiv 2014, 1406.1231, https://arxiv.org/abs/1406.1231. · arXiv
- 10.1021/acs.jcim.8b00785 · DOI · Paper page
- Ramsundar B., 2015, arXiv
- McMahan H. B., 2017, Proc. 20th Int. Conf. Artif. Intell. Stat. AISTATS 2017, 54, 1273
- 10.1093/nar/gkw1074 · DOI · Paper page
- 10.1016/bs.armc.2017.08.004 · DOI
- 10.1021/ci100050t · DOI · Paper page
- Arany, A.; Simm, J.; Oldenhof, M.; Moreau, Y. SparseChem: Fast and Accurate Machine Learning Model for Small Molecules. arXiv 2022, 2203.04676, https://arxiv.org/abs/2203.04676. · arXiv
- Galtier, M.; Marini, C. Substra: A Framework for Privacy-Preserving, Traceable and Collaborative Machine Learning. arXiv 2019, 1910.11567, https://arxiv.org/abs/1910.11567. · arXiv
- 10.1609/aaai.v37i13.26847 · DOI
- Wright S., 1921, J. Agric. Res., 557
- 10.1186/s13321-021-00576-2 · DOI
- 10.1186/s13321-017-0230-2 · DOI
- 10.1021/ci9000579 · DOI · Paper page
- 10.1021/ci5001168 · DOI · Paper page
- 10.1016/j.ailsci.2023.100070 · DOI
- 10.5555/3295222.3295387 · doi:10.5555/3295222.3295387
- 10.1021/acs.jcim.6b00572 · DOI
- 10.1186/s13321-021-00487-2 · DOI
- 10.1093/bioinformatics/btaa1006 · DOI · Paper page
- Smith V., 2017, Advances in Neural Information Processing Systems, 30
- 10.1093/bioinformatics/btaa038 · DOI
- 10.1021/acs.jcim.0c01342 · DOI
- 10.1186/s13321-018-0281-z · DOI
- 10.1002/(SICI)1097-0290(199824)61:1<47::AID-BIT9>3.0.CO;2-Z · DOI · Paper page
- 10.1021/ci00046a002 · DOI · Paper page
- 10.1021/ci00054a008 · DOI · Paper page
- Gilmer J., 2017, 34th Int. Conf. Mach. Learn. ICML 2017, 3, 2053
- Vaswani A., 2017, Advances in Neural Information Processing Systems, 30
- 10.1039/C8SC04175J · DOI · Paper page
Cited by
Teo, Z. L., Jin, L., Liu, N., Li, S., Miao, D., Zhang, X., Ng, W. Y., Tan, T. F., Lee, D. M., Chua, K. J., et al. (2024) · Cell Reports Medicine · doi:10.1016/j.xcrm.2024.101419 · Cited by 255
Unleashing the power of generative AI in drug discovery
Gangwal, A., & Lavecchia, A. (2024) · Drug Discovery Today · doi:10.1016/j.drudis.2024.103992 · Cited by 195
Recent advances in AI-based toxicity prediction for drug discovery
Lee, H., Kim, J., Kim, J.-W., & Lee, Y. (2025) · Frontiers in Chemistry · doi:10.3389/fchem.2025.1632046 · Cited by 76
Zhao, J., Bagchi, S., Avestimehr, S., Chan, K., Chaterji, S., Dimitriadis, D., Li, J., Li, N., Nourian, A., & Roth, H. (2025) · ACM Computing Surveys · doi:10.1145/3724113 · Cited by 63
Carratt, S. A., Zuch de Zafra, C. L., Oziolor, E., Rana, P., Vansell, N. R., Mangipudy, R., & Vaidya, V. S. (2025) · Toxicological Sciences · doi:10.1093/toxsci/kfae122 · Cited by 59
Metadata
- Authors
- Heyndrickx, W., Mervin, L., Morawietz, T., Sturm, N., Friedrich, L., Zalewski, A., Pentina, A., Humbeck, L., Oldenhof, M., Niwayama, R., Schmidtke, P., Fechner, N., Simm, J., Arany, A., Drizard, N., Jabal, R., Afanasyeva, A., Loeb, R., Verma, S., Harnqvist, S., Holmes, M., Pejo, B., Telenczuk, M., Holway, N., Dieckmann, A., Rieke, N., Zumsande, F., Clevert, D.-A., Krug, M., Luscombe, C., Green, D., Ertl, P., Antal, P., Marcus, D., Do Huu, N., Fuji, H., Pickett, S., Acs, G., Boniface, E., Beck, B., Sun, Y., Gohier, A., Rippmann, F., Engkvist, O., Göller, A. H., Moreau, Y., Galtier, M. N., Schuffenhauer, A., & Ceulemans, H.
- Venue
- Journal of Chemical Information and Modeling
- Publisher
- American Chemical Society (ACS)
- Published
- 2023-08-29
- Volume
- 64
- Issue
- 7
- Pages
- 2331-2344
- ISSN
- 1549-9596 (ISSN-L)
- 1549-960X
- OpenAlex
- W4386256707
- Cited by
- 170
- References
- 19
- FWCI
- 22.37
- Type
- journal article
- Language
- en
- Open access
- hybrid
- Funders
- KU Leuven
- Innovative Medicines Initiative (831472)
- Horizon 2020 Framework Programme
- European Federation of Pharmaceutical Industries and Associations
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.