{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,15]],"date-time":"2026-10-15T23:28:03Z","timestamp":1792106883781,"version":"4.4.0"},"reference-count":43,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T00:00:00Z","timestamp":1778198400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,7,15]],"date-time":"2026-07-15T00:00:00Z","timestamp":1784073600000},"content-version":"vor","delay-in-days":68,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100014013","name":"UK Research and Innovation","doi-asserted-by":"publisher","award":["BB\/W013959\/1"],"award-info":[{"award-number":["BB\/W013959\/1"]}],"id":[{"id":"10.13039\/100014013","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["npj Syst Biol Appl"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Gene essentiality, the requirement of a gene for survival or proliferation, is central to understanding cellular processes and identifying drug targets. Experimental determination requires large growth screens that are time-consuming and expensive, motivating in silico approaches. Existing methods predominantly use flux balance analysis (FBA), a constraint-based optimisation framework that requires a predefined cellular objective function. This can introduce observer bias, because the objective often reflects the researcher\u2019s assumptions rather than the cell\u2019s biological goals. Here, we present FluxGAT, a graph neural network (GNN) that predicts gene essentiality from graphical representations of flux sampling data. Flux sampling removes the need for an explicit objective and instead characterises feasible steady-state fluxes. FluxGAT combines this information with metabolic network topology to learn flux-informed node representations and classify reactions as essential or non-essential. We apply FluxGAT to the iCHO2291 genome-scale model of Chinese hamster ovary cells and Mouse1, a generic mouse model with independent essentiality labels. In both systems, FluxGAT improves sensitivity over FBA while maintaining high precision and specificity, and recovers more experimentally essential genes, especially where FBA predicts very few essentials. These results show that flux-informed GNNs can provide more general gene essentiality predictions across mammalian genome-scale models without hand-crafted objective functions.<\/jats:p>","DOI":"10.1038\/s41540-026-00738-8","type":"journal-article","created":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T15:03:59Z","timestamp":1778252639000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Flux sampling and graph neural networks for improved gene essentiality prediction in mammalian genome-scale metabolic models"],"prefix":"10.1038","volume":"12","author":[{"given":"Kieren","family":"Sharma","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lucia","family":"Marucci","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zahraa S.","family":"Abdallah","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"297","published-online":{"date-parts":[[2026,5,8]]},"reference":[{"key":"738_CR1","doi-asserted-by":"publisher","first-page":"34","DOI":"10.1038\/nrg.2017.74","volume":"19","author":"G Rancati","year":"2018","unstructured":"Rancati, G., Moffat, J., Typas, A. & Pavelka, N. Emerging and evolving concepts in gene essentiality. Nat. Rev. Genet. 19, 34\u201349 (2018).","journal-title":"Nat. Rev. Genet."},{"key":"738_CR2","doi-asserted-by":"publisher","first-page":"e45379","DOI":"10.7554\/eLife.45379","volume":"8","author":"J-C Lachance","year":"2019","unstructured":"Lachance, J.-C., Rodrigue, S. & Palsson, B. O. Minimal cells, maximal knowledge. Elife 8, e45379 (2019).","journal-title":"Elife"},{"key":"738_CR3","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-020-14284-2","volume":"11","author":"P Cacheiro","year":"2020","unstructured":"Cacheiro, P. et al. Human and mouse essentiality screens as a resource for disease gene discovery. Nat. Commun. 11, 655 (2020).","journal-title":"Nat. Commun."},{"key":"738_CR4","doi-asserted-by":"publisher","first-page":"544","DOI":"10.1016\/j.drudis.2018.11.002","volume":"24","author":"X Ji","year":"2019","unstructured":"Ji, X., Rajpal, D. K. & Freudenberg, J. M. The essentiality of drug targets: an analysis of current literature and genomic databases. Drug Discov. Today 24, 544\u2013550 (2019).","journal-title":"Drug Discov. Today"},{"key":"738_CR5","doi-asserted-by":"publisher","first-page":"4579","DOI":"10.1016\/j.cell.2021.06.033","volume":"184","author":"B Bosch","year":"2021","unstructured":"Bosch, B. et al. Genome-wide gene expression tuning reveals diverse vulnerabilities of m. tuberculosis. Cell 184, 4579\u20134592 (2021).","journal-title":"Cell"},{"key":"738_CR6","doi-asserted-by":"publisher","unstructured":"Wang, W. et al. A gene essentiality signature for studying the mechanism of action of drugs. Preprint at https:\/\/doi.org\/10.1101\/2022.11.07.514541 (2022).","DOI":"10.1101\/2022.11.07.514541"},{"key":"738_CR7","doi-asserted-by":"publisher","first-page":"466","DOI":"10.1016\/j.ccell.2020.12.008","volume":"39","author":"L Chang","year":"2021","unstructured":"Chang, L., Ruiz, P., Ito, T. & Sellers, W. R. Targeting pan-essential genes in cancer: challenges and opportunities. Cancer Cell 39, 466\u2013479 (2021).","journal-title":"Cancer Cell"},{"key":"738_CR8","doi-asserted-by":"publisher","first-page":"492","DOI":"10.1021\/acssynbio.2c00481","volume":"12","author":"K Niu","year":"2023","unstructured":"Niu, K. et al. High-level production of l-methionine by dynamic deregulation of metabolism with engineered nonauxotrophic Escherichia coli. ACS Synth. Biol. 12, 492\u2013501 (2023).","journal-title":"ACS Synth. Biol."},{"key":"738_CR9","doi-asserted-by":"publisher","first-page":"566","DOI":"10.1093\/bib\/bbz017","volume":"21","author":"X Li","year":"2020","unstructured":"Li, X., Li, W., Zeng, M., Zheng, R. & Li, M. Network-based methods for predicting essential genes or proteins: a survey. Brief. Bioinformatics 21, 566\u2013583 (2020).","journal-title":"Brief. Bioinformatics"},{"key":"738_CR10","doi-asserted-by":"publisher","first-page":"bbab128","DOI":"10.1093\/bib\/bbab128","volume":"22","author":"O Aromolaran","year":"2021","unstructured":"Aromolaran, O., Aromolaran, D., Isewon, I. & Oyelade, J. Machine learning approach to gene essentiality prediction: a review. Brief. Bioinformatics 22, bbab128 (2021).","journal-title":"Brief. Bioinformatics"},{"key":"738_CR11","doi-asserted-by":"publisher","first-page":"13","DOI":"10.1016\/j.ifacol.2023.01.006","volume":"55","author":"LJ Freischem","year":"2022","unstructured":"Freischem, L. J., Barahona, M. & Oyarz\u00fan, D. A. Prediction of gene essentiality using machine learning and genome-scale metabolic models. IFAC-PapersOnLine 55, 13\u201318 (2022).","journal-title":"IFAC-PapersOnLine"},{"key":"738_CR12","doi-asserted-by":"publisher","first-page":"442","DOI":"10.1016\/j.tig.2020.03.005","volume":"36","author":"CB Azodi","year":"2020","unstructured":"Azodi, C. B., Tang, J. & Shiu, S.-H. Opening the black box: interpretable machine learning for geneticists. Trends Genet. 36, 442\u2013455 (2020).","journal-title":"Trends Genet."},{"key":"738_CR13","doi-asserted-by":"crossref","unstructured":"Wu, L., Cui, P., Pei, J., Zhao, L. & Guo, X. Graph neural networks: foundation, frontiers and applications. In Proc. 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, 4840\u20134841 (ACM, 2022).","DOI":"10.1145\/3534678.3542609"},{"key":"738_CR14","doi-asserted-by":"publisher","first-page":"9963","DOI":"10.1007\/s00521-024-09662-6","volume":"36","author":"K Lazaros","year":"2024","unstructured":"Lazaros, K., Koumadorakis, D. E., Vlamos, P. & Vrahatis, A. G. Graph neural network approaches for single-cell data: a recent overview. Neural Comput. Appl. 36, 9963\u20139987 (2024).","journal-title":"Neural Comput. Appl."},{"key":"738_CR15","doi-asserted-by":"publisher","first-page":"245","DOI":"10.1038\/nbt.1614","volume":"28","author":"JD Orth","year":"2010","unstructured":"Orth, J. D., Thiele, I. & Palsson, B. \u00d8 What is flux balance analysis? Nat. Biotechnol. 28, 245\u2013248 (2010).","journal-title":"Nat. Biotechnol."},{"key":"738_CR16","doi-asserted-by":"publisher","first-page":"119","DOI":"10.1038\/msb4100162","volume":"3","author":"R Schuetz","year":"2007","unstructured":"Schuetz, R., Kuepfer, L. & Sauer, U. Systematic evaluation of objective functions for predicting intracellular fluxes in Escherichia coli. Mol. Syst. Biol. 3, 119 (2007).","journal-title":"Mol. Syst. Biol."},{"key":"738_CR17","doi-asserted-by":"publisher","first-page":"985","DOI":"10.1002\/btpr.1949","volume":"30","author":"CE Garc\u00eda S\u00e1nchez","year":"2014","unstructured":"Garc\u00eda S\u00e1nchez, C. E. & Torres S\u00e1ez, R. G. Comparison and analysis of objective functions in flux balance analysis. Biotechnol. Prog. 30, 985\u2013991 (2014).","journal-title":"Biotechnol. Prog."},{"key":"738_CR18","doi-asserted-by":"publisher","first-page":"e0276112","DOI":"10.1371\/journal.pone.0276112","volume":"17","author":"B Schnitzer","year":"2022","unstructured":"Schnitzer, B., \u00d6sterberg, L. & Cvijovic, M. The choice of the objective function in flux balance analysis is crucial for predicting replicative lifespans in yeast. Plos ONE 17, e0276112 (2022).","journal-title":"Plos ONE"},{"key":"738_CR19","doi-asserted-by":"publisher","first-page":"e1009589","DOI":"10.1371\/journal.pcbi.1009589","volume":"17","author":"S Ravi","year":"2021","unstructured":"Ravi, S. & Gunawan, R. \u03b4fba\u2013predicting metabolic flux alterations using genome-scale metabolic models and differential transcriptomic data. PLoS Comput. Biol. 17, e1009589 (2021).","journal-title":"PLoS Comput. Biol."},{"key":"738_CR20","doi-asserted-by":"publisher","DOI":"10.1038\/s41540-019-0109-0","volume":"5","author":"HA Herrmann","year":"2019","unstructured":"Herrmann, H. A., Dyson, B. C., Vass, L., Johnson, G. N. & Schwartz, J.-M. Flux sampling is a powerful tool to study metabolism under changing environmental conditions. npj Syst. Biol. Appl.5, 32 (2019).","journal-title":"npj Syst. Biol. Appl."},{"key":"738_CR21","doi-asserted-by":"crossref","unstructured":"Strain, B., Morrissey, J., Antonakoudis, A. & Kontoravdi, C. How reliable are Chinese hamster ovary (CHO) cell genome-scale metabolic models?Biotechnol. Bioeng. 120, 2460\u20132478 (2023).","DOI":"10.1002\/bit.28366"},{"key":"738_CR22","doi-asserted-by":"publisher","DOI":"10.1186\/s12859-024-05655-3","volume":"25","author":"PE Gelbach","year":"2024","unstructured":"Gelbach, P. E., Cetin, H. & Finley, S. D. Flux sampling in genome-scale metabolic modeling of microbial communities. BMC Bioinformatics 25, 45 (2024).","journal-title":"BMC Bioinformatics"},{"key":"738_CR23","doi-asserted-by":"publisher","DOI":"10.1038\/s41540-024-00348-2","volume":"10","author":"R Hasibi","year":"2024","unstructured":"Hasibi, R., Michoel, T. & Oyarz\u00fan, D. A. Integration of graph neural networks and genome-scale metabolic models for predicting gene essentiality. npj Syst. Biol. Appl. 10, 24 (2024).","journal-title":"npj Syst. Biol. Appl."},{"key":"738_CR24","doi-asserted-by":"publisher","first-page":"1615","DOI":"10.1109\/TCBB.2021.3054738","volume":"19","author":"J Schapke","year":"2021","unstructured":"Schapke, J., Tavares, A. & Recamonde-Mendoza, M. Epgat: gene essential- ity prediction with graph attention networks. IEEE\/ACM Transactions on Computational Biology and Bioinformatics 19, 1615\u20131626 (2021).","journal-title":"IEEE\/ACM Transactions on Computational Biology and Bioinformatics"},{"key":"738_CR25","doi-asserted-by":"publisher","first-page":"138","DOI":"10.1016\/j.ymben.2020.04.005","volume":"60","author":"HC Yeo","year":"2020","unstructured":"Yeo, H. C., Hong, J., Lakshmanan, M. & Lee, D.-Y. Enzyme capacity-based genome scale modelling of cho cells. Metab. Eng. 60, 138\u2013147 (2020).","journal-title":"Metab. Eng."},{"key":"738_CR26","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-74735-0","volume":"10","author":"N Yamano-Adachi","year":"2020","unstructured":"Yamano-Adachi, N., Arishima, R., Puriwat, S. & Omasa, T. Establishment of fast-growing serum-free immortalised cells from Chinese hamster lung tissues for biopharmaceutical production. Sci. Rep. 10, 17612 (2020).","journal-title":"Sci. Rep."},{"key":"738_CR27","doi-asserted-by":"publisher","DOI":"10.1038\/s41540-018-0067-y","volume":"4","author":"M Beguerisse-D\u00edaz","year":"2018","unstructured":"Beguerisse-D\u00edaz, M., Bosque, G., Oyarz\u00fan, D., Pic\u00f3, J. & Barahona, M. Flux-dependent graphs for metabolic networks. NPJ Syst. Biol. Appl. 4, 32 (2018).","journal-title":"NPJ Syst. Biol. Appl."},{"key":"738_CR28","doi-asserted-by":"crossref","unstructured":"Galuzzi, B. G., Milazzo, L. & Damiani, C. Best practices in flux sampling of constrained-based models. In Proc. International Conference on Machine Learning, Optimization, and Data Science 234\u2013248 (Springer, 2022).","DOI":"10.1007\/978-3-031-25891-6_18"},{"key":"738_CR29","doi-asserted-by":"crossref","unstructured":"Xiong, K. et al. An optimized genome-wide, virus-free CRISPR screen for mammalian cells. Cell Rep. Methods 1, 100062 (2021).","DOI":"10.1016\/j.crmeth.2021.100062"},{"key":"738_CR30","doi-asserted-by":"publisher","unstructured":"Loshchilov, I. & Hutter, F. Decoupled weight decay regularization. Prerpint at https:\/\/doi.org\/10.48550\/arXiv.1711.05101 (2017).","DOI":"10.48550\/arXiv.1711.05101"},{"key":"738_CR31","doi-asserted-by":"crossref","unstructured":"Grover, A. & Leskovec, J. node2vec: scalable feature learning for networks. In (ed.) Proc. 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 855\u2013864 (ACM, 2016).","DOI":"10.1145\/2939672.2939754"},{"key":"738_CR32","doi-asserted-by":"publisher","DOI":"10.1186\/1752-0509-2-50","volume":"2","author":"D Deutscher","year":"2008","unstructured":"Deutscher, D., Meilijson, I., Schuster, S. & Ruppin, E. Can single knockouts accurately single out gene functions? BMC Syst. Biol. 2, 50 (2008).","journal-title":"BMC Syst. Biol."},{"key":"738_CR33","doi-asserted-by":"publisher","first-page":"3299","DOI":"10.1093\/bioinformatics\/btv352","volume":"31","author":"A Pratapa","year":"2015","unstructured":"Pratapa, A., Balachandran, S. & Raman, K. Fast-sl: an efficient algorithm to identify synthetic lethal sets in metabolic networks. Bioinformatics 31, 3299\u20133305 (2015).","journal-title":"Bioinformatics"},{"key":"738_CR34","doi-asserted-by":"publisher","first-page":"e2102344118","DOI":"10.1073\/pnas.2102344118","volume":"118","author":"H Wang","year":"2021","unstructured":"Wang, H. et al. Genome-scale metabolic network reconstruction of model animals as a platform for translational research. Proc. Natl. Acad. Sci. USA 118, e2102344118 (2021).","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"738_CR35","doi-asserted-by":"crossref","unstructured":"Chen, W.-H., Lu, G., Chen, X., Zhao, X.-M. & Bork, P. Ogee v2: an update of the online gene essentiality database with special focus on differentially essential genes in human cancer cell lines. Nucleic Acids Res. 45, D940\u2013D944 (2016).","DOI":"10.1093\/nar\/gkw1013"},{"key":"738_CR36","doi-asserted-by":"publisher","first-page":"W207","DOI":"10.1093\/nar\/gkad347","volume":"51","author":"L Kolberg","year":"2023","unstructured":"Kolberg, L. et al. g: Profiler\u2013interoperable web service for functional enrichment analysis and gene identifier mapping (2023 update). Nucleic Acids Res. 51, W207\u2013W212 (2023).","journal-title":"Nucleic Acids Res."},{"key":"738_CR37","doi-asserted-by":"publisher","first-page":"D1038","DOI":"10.1093\/nar\/gkac972","volume":"51","author":"T Groza","year":"2023","unstructured":"Groza, T. et al. The international mouse phenotyping consortium: comprehensive knockout phenotyping underpinning the study of human disease. Nucleic Acids Res. 51, D1038\u2013D1045 (2023).","journal-title":"Nucleic Acids Res."},{"key":"738_CR38","doi-asserted-by":"publisher","first-page":"15112","DOI":"10.1073\/pnas.232349399","volume":"99","author":"D Segre","year":"2002","unstructured":"Segre, D., Vitkup, D. & Church, G. M. Analysis of optimality in natural and perturbed metabolic networks. Proc. Natl. Acad. Sci. USA 99, 15112\u201315117 (2002).","journal-title":"Proc. Natl. Acad. Sci. USA"},{"key":"738_CR39","doi-asserted-by":"publisher","first-page":"e0235393","DOI":"10.1371\/journal.pone.0235393","volume":"15","author":"S Fallahi","year":"2020","unstructured":"Fallahi, S., Skaug, H. J. & Alendal, G. A comparison of monte carlo sampling methods for metabolic network models. Plos ONE 15, e0235393 (2020).","journal-title":"Plos ONE"},{"key":"738_CR40","doi-asserted-by":"publisher","first-page":"e86587","DOI":"10.1371\/journal.pone.0086587","volume":"9","author":"W Megchelenbrink","year":"2014","unstructured":"Megchelenbrink, W., Huynen, M. & Marchiori, E. optgpsampler: an improved tool for uniformly sampling the solution-space of genome-scale metabolic networks. PloS ONE 9, e86587 (2014).","journal-title":"PloS ONE"},{"key":"738_CR41","doi-asserted-by":"publisher","first-page":"e1000859","DOI":"10.1371\/journal.pcbi.1000859","volume":"6","author":"S Bordel","year":"2010","unstructured":"Bordel, S., Agren, R. & Nielsen, J. Sampling the solution space in genome-scale metabolic networks reveals transcriptional regulation in key enzymes. PLoS Comput. Biol. 6, e1000859 (2010).","journal-title":"PLoS Comput. Biol."},{"key":"738_CR42","doi-asserted-by":"publisher","first-page":"1986","DOI":"10.1093\/bioinformatics\/btz832","volume":"36","author":"M Ponce-de Le\u00f3n","year":"2020","unstructured":"Ponce-de Le\u00f3n, M., Apaolaza, I., Valencia, A. & Planes, F. J. On the inconsistent treatment of gene-protein-reaction rules in context-specific metabolic models. Bioinformatics 36, 1986 (2020).","journal-title":"Bioinformatics"},{"key":"738_CR43","doi-asserted-by":"publisher","unstructured":"Veli\u010dkovi\u0107, P. et al. Graph attention networks. Preprint at https:\/\/doi.org\/10.48550\/arXiv.1710.10903 (2017).","DOI":"10.48550\/arXiv.1710.10903"}],"container-title":["npj Systems Biology and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.nature.com\/articles\/s41540-026-00738-8","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41540-026-00738-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/www.nature.com\/articles\/s41540-026-00738-8.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,14]],"date-time":"2026-07-14T23:48:12Z","timestamp":1784072892000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.nature.com\/articles\/s41540-026-00738-8"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,8]]},"references-count":43,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,12]]}},"alternative-id":["738"],"URL":"https:\/\/doi.org\/10.1038\/s41540-026-00738-8","relation":{},"ISSN":["2056-7189"],"issn-type":[{"value":"2056-7189","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,8]]},"assertion":[{"value":"19 October 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"29 April 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"8 May 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"The authors declare no competing interests.","order":1,"name":"Ethics","label":"Competing interests","group":{"name":"EthicsHeading","label":"Ethics"}}],"article-number":"107"}}