{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,30]],"date-time":"2026-10-30T04:17:20Z","timestamp":1793333840923,"version":"4.5.0"},"update-to":[{"DOI":"10.1371\/journal.pcbi.1013312","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T00:00:00Z","timestamp":1773964800000}}],"reference-count":41,"publisher":"Public Library of Science (PLoS)","issue":"3","license":[{"start":{"date-parts":[[2026,3,16]],"date-time":"2026-03-16T00:00:00Z","timestamp":1773619200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["www.ploscompbiol.org"],"crossmark-restriction":false},"short-container-title":["PLoS Comput Biol"],"abstract":"<jats:p>We present a generative modeling framework for global sensitivity analysis (GSA) in complex systems characterized by strong and potentially high-dimensional parameter correlations. Traditional variance-based GSA methods rely on the assumption of independent inputs, which rarely holds for Bayesian-calibrated models. While recent extensions using Rosenblatt transformations and Shapley effects theoretically address this limitation, their implementation requires accurate conditional sampling from correlated joint distributions, a task that remains challenging. Existing solutions suffer from restrictive assumptions on input dependence, which limit their applicability to complex data-driven problems. Our method addresses these challenges by reframing sensitivity analysis as a post calibration task on Bayesian posterior distributions, where parameter correlations are learned from data using generative models, eliminating restrictive dependence assumptions and ensuring data relevant sensitivity estimates. We employ autoregressive architectures to implement Rosenblatt transformations and leverage diffusion models to estimate Shapley effects. These methods impose no predefined distributional assumptions and scale efficiently with both data volume and model complexity. We demonstrate the effectiveness of our approach on two representative applications: a COVID-19 transmission model and a cancer immunotherapy model. Results show that our methods effectively captures parameter sensitivities in the presence of parameter correlations, and achieve notable gains in scalability and flexibility over existing methods.<\/jats:p>","DOI":"10.1371\/journal.pcbi.1013312","type":"journal-article","created":{"date-parts":[[2026,3,16]],"date-time":"2026-03-16T17:40:25Z","timestamp":1773682825000},"page":"e1013312","update-policy":"https:\/\/doi.org\/10.1371\/journal.pcbi.corrections_policy","source":"Crossref","is-referenced-by-count":2,"title":["Bayesian-calibrated global sensitivity analysis for mathematical models using generative AI"],"prefix":"10.1371","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-4470-8200","authenticated-orcid":true,"given":"Xuyuan","family":"Wang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"340","published-online":{"date-parts":[[2026,3,16]]},"reference":[{"issue":"2","key":"pcbi.1013312.ref001","doi-asserted-by":"crossref","first-page":"109","DOI":"10.1093\/bib\/bbm007","article-title":"Bayesian methods in bioinformatics and computational systems biology","volume":"8","author":"DJ Wilkinson","year":"2007","journal-title":"Brief Bioinform"},{"key":"pcbi.1013312.ref002","doi-asserted-by":"crossref","first-page":"271","DOI":"10.1016\/S0378-4754(00)00270-6","article-title":"Global sensitivity indices for nonlinear mathematical models and their Monte Carlo estimates","volume":"55","author":"IM Sobol\u2032","year":"2001","journal-title":"Mathematics and Computers in Simulation"},{"key":"pcbi.1013312.ref003","doi-asserted-by":"crossref","first-page":"107300","DOI":"10.1016\/j.ress.2020.107300","article-title":"Variance-based sensitivity analysis: The quest for better estimators and designs between explorativity and economy","volume":"206","author":"S Lo Piano","year":"2021","journal-title":"Reliability Engineering & System Safety"},{"key":"pcbi.1013312.ref004","doi-asserted-by":"crossref","first-page":"108306","DOI":"10.1016\/j.mbs.2020.108306","article-title":"Sensitivity analysis methods in the biomedical sciences","volume":"323","author":"G Qian","year":"2020","journal-title":"Math Biosci"},{"issue":"86","key":"pcbi.1013312.ref005","doi-asserted-by":"crossref","first-page":"20121018","DOI":"10.1098\/rsif.2012.1018","article-title":"Sensitivity analysis of infectious disease models: methods, advances and their application","volume":"10","author":"J Wu","year":"2013","journal-title":"J R Soc Interface"},{"key":"pcbi.1013312.ref006","doi-asserted-by":"crossref","first-page":"173","DOI":"10.1016\/j.envsoft.2015.07.010","article-title":"Non-parametric methods for global sensitivity analysis of model output with dependent inputs","volume":"72","author":"TA Mara","year":"2015","journal-title":"Environmental Modelling & Software"},{"issue":"1","key":"pcbi.1013312.ref007","doi-asserted-by":"crossref","first-page":"1060","DOI":"10.1137\/15M1048070","article-title":"Shapley Effects for Global Sensitivity Analysis: Theory and Computation","volume":"4","author":"E Song","year":"2016","journal-title":"SIAM\/ASA J Uncertainty Quantification"},{"key":"pcbi.1013312.ref008","doi-asserted-by":"crossref","first-page":"107811","DOI":"10.1016\/j.ast.2022.107811","article-title":"Copula-based methods for global sensitivity analysis with correlated random variables and stochastic processes under incomplete probability information","volume":"129","author":"S Song","year":"2022","journal-title":"Aerospace Science and Technology"},{"issue":"2","key":"pcbi.1013312.ref009","first-page":"182","article-title":"Pair-copula constructions of multiple dependence","volume":"44","author":"K Aas","year":"2009","journal-title":"Insurance: Mathematics and Economics"},{"issue":"11","key":"pcbi.1013312.ref010","doi-asserted-by":"crossref","first-page":"7327","DOI":"10.1109\/TPAMI.2021.3116668","article-title":"Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Models","volume":"44","author":"S Bond-Taylor","year":"2022","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"2","key":"pcbi.1013312.ref011","doi-asserted-by":"crossref","first-page":"308","DOI":"10.1016\/0021-9991(85)90148-2","article-title":"Application of a fractional-step method to incompressible Navier-Stokes equations","volume":"59","author":"J Kim","year":"1985","journal-title":"Journal of Computational Physics"},{"issue":"220","key":"pcbi.1013312.ref012","doi-asserted-by":"crossref","first-page":"20240394","DOI":"10.1098\/rsif.2024.0394","article-title":"The 1978 English boarding school influenza outbreak: where the classic SEIR model fails","volume":"21","author":"KK Avilov","year":"2024","journal-title":"J R Soc Interface"},{"key":"pcbi.1013312.ref013","first-page":"221","article-title":"Bayesian inference for dynamical systems","volume":"5","author":"WC Roda","year":"2020","journal-title":"Infect Dis Model"},{"issue":"6","key":"pcbi.1013312.ref014","doi-asserted-by":"crossref","first-page":"7745","DOI":"10.1016\/j.sbspro.2010.05.208","article-title":"Derivative based global sensitivity measures","volume":"2","author":"IM Sobol","year":"2010","journal-title":"Procedia - Social and Behavioral Sciences"},{"key":"pcbi.1013312.ref015","doi-asserted-by":"crossref","unstructured":"Blower SM, Dowlatabadi H. Sensitivity and uncertainty analysis of complex models of disease transmission: an HIV model, as an example. International Statistical Review\/Revue Internationale de Statistique. 1994;:229\u201343.","DOI":"10.2307\/1403510"},{"key":"pcbi.1013312.ref016","volume-title":"Global sensitivity analysis: the primer","author":"A Saltelli","year":"2008"},{"issue":"7","key":"pcbi.1013312.ref017","doi-asserted-by":"crossref","first-page":"957","DOI":"10.1016\/j.ress.2006.07.001","article-title":"Estimating the approximation error when fixing unessential factors in global sensitivity analysis","volume":"92","author":"IM Sobol\u2019","year":"2007","journal-title":"Reliability Engineering & System Safety"},{"key":"pcbi.1013312.ref018","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.compag.2013.10.006","article-title":"Global sensitivity analysis by means of EFAST and Sobol\u2019 methods and calibration of reduced state-variable TOMGRO model using genetic algorithms","volume":"100","author":"MA Vazquez-Cruz","year":"2014","journal-title":"Computers and Electronics in Agriculture"},{"issue":"1","key":"pcbi.1013312.ref019","doi-asserted-by":"crossref","first-page":"72","DOI":"10.1287\/ijoc.10.1.72","article-title":"Numerical Methods for Fitting and Simulating Autoregressive-to-Anything Processes","volume":"10","author":"MC Cario","year":"1998","journal-title":"INFORMS Journal on Computing"},{"key":"pcbi.1013312.ref020","doi-asserted-by":"crossref","unstructured":"Scott EM. Uncertainty and sensitivity studies of models of environmental systems. In: Proceedings of the 28th conference on Winter simulation - WSC \u201996, 1996. 255\u20139. https:\/\/doi.org\/10.1145\/256562.256618","DOI":"10.1145\/256562.256618"},{"issue":"3","key":"pcbi.1013312.ref021","doi-asserted-by":"crossref","first-page":"276","DOI":"10.1145\/937332.937336","article-title":"Behavior of the NORTA method for correlated random vector generation as the dimension increases","volume":"13","author":"S Ghosh","year":"2003","journal-title":"ACM Trans Model Comput Simul"},{"issue":"2","key":"pcbi.1013312.ref022","doi-asserted-by":"crossref","first-page":"93","DOI":"10.1080\/00031305.1998.10480547","article-title":"Markov Chain Monte Carlo in Practice: A Roundtable Discussion","volume":"52","author":"RE Kass","year":"1998","journal-title":"The American Statistician"},{"issue":"1","key":"pcbi.1013312.ref023","doi-asserted-by":"crossref","first-page":"261","DOI":"10.1111\/j.2517-6161.1994.tb01976.x","article-title":"On the Convergence of Monte Carlo Maximum Likelihood Calculations","volume":"56","author":"CJ Geyer","year":"1994","journal-title":"Journal of the Royal Statistical Society Series B: Statistical Methodology"},{"issue":"1","key":"pcbi.1013312.ref024","doi-asserted-by":"crossref","DOI":"10.1214\/10-AOS834","article-title":"Consistency of the maximum likelihood estimator for general hidden Markov models","volume":"39","author":"R Douc","year":"2011","journal-title":"Ann Statist"},{"issue":"2","key":"pcbi.1013312.ref025","doi-asserted-by":"crossref","first-page":"916","DOI":"10.1093\/gji\/ggac100","article-title":"Upscaling and downscaling Monte Carlo ensembles with generative models","volume":"230","author":"M Scheiter","year":"2022","journal-title":"Geophysical Journal International"},{"key":"pcbi.1013312.ref026","unstructured":"Germain M, Gregor K, Murray I, Larochelle H. MADE: masked autoencoder for distribution estimation. In: Proceedings of the 32nd International Conference on Machine Learning, 2015. 881\u20139."},{"key":"pcbi.1013312.ref027","unstructured":"Papamakarios G, Pavlakou T, Murray I. Masked autoregressive flow for density estimation. 2018. https:\/\/arxiv.org\/abs\/1705.07057"},{"key":"pcbi.1013312.ref028","unstructured":"Karpathy A. Pytorch-made: MADE (Masked Autoencoder Density Estimation) implementation in PyTorch. https:\/\/github.com\/karpathy\/pytorch-made. 2025."},{"key":"pcbi.1013312.ref029","author":"X WANG","year":"2025","journal-title":"JRS_code"},{"key":"pcbi.1013312.ref030","unstructured":"Ho J, Jain A, Abbeel P. Denoising diffusion probabilistic models. 2020. https:\/\/arxiv.org\/abs\/2006.11239"},{"key":"pcbi.1013312.ref031","unstructured":"Ho J, Salimans T. Classifier-free diffusion guidance. 2022. https:\/\/arxiv.org\/abs\/2207.12598"},{"key":"pcbi.1013312.ref032","doi-asserted-by":"crossref","unstructured":"Lugmayr A, Danelljan M, Romero A, Yu F, Timofte R, Gool LV. RePaint: inpainting using denoising diffusion probabilistic models. 2022. https:\/\/arxiv.org\/abs\/2201.09865","DOI":"10.1109\/CVPR52688.2022.01117"},{"issue":"7","key":"pcbi.1013312.ref033","doi-asserted-by":"crossref","first-page":"964","DOI":"10.1016\/j.ress.2007.04.002","article-title":"Global sensitivity analysis using polynomial chaos expansions","volume":"93","author":"B Sudret","year":"2008","journal-title":"Reliability Engineering & System Safety"},{"key":"pcbi.1013312.ref034","doi-asserted-by":"crossref","unstructured":"Marelli S, Sudret B. UQLab: A framework for uncertainty quantification in Matlab. Vulnerability, uncertainty, and risk: quantification, mitigation, and management. 2014. p. 2554\u201363.","DOI":"10.1061\/9780784413609.257"},{"key":"pcbi.1013312.ref035","doi-asserted-by":"crossref","unstructured":"Iooss B, Prieur C. Shapley effects for sensitivity analysis with correlated inputs: comparisons with Sobol indices, numerical estimation and applications; 2019. Available from: https:\/\/arxiv.org\/abs\/1707.01334","DOI":"10.1615\/Int.J.UncertaintyQuantification.2019028372"},{"key":"pcbi.1013312.ref036","article-title":"Benchmark of goal oriented sensitivity analysis methods using Ishigami function","volume":"3","author":"Z Kala","year":"2018","journal-title":"International Journal of Mathematical and Computational Methods"},{"issue":"1","key":"pcbi.1013312.ref037","first-page":"112","article-title":"On sensitivity estimation for nonlinear mathematical models","volume":"2","author":"IM Sobol","year":"1990","journal-title":"Matematicheskoe modelirovanie"},{"key":"pcbi.1013312.ref038","unstructured":"China CDC. Tracking the Epidemic. https:\/\/weekly.chinacdc.cn\/news\/TrackingtheEpidemic.htm. 2020."},{"issue":"22","key":"pcbi.1013312.ref039","doi-asserted-by":"crossref","first-page":"5576","DOI":"10.3390\/cancers14225576","article-title":"Modeling Patient-Specific CAR-T Cell Dynamics: Multiphasic Kinetics via Phenotypic Differentiation","volume":"14","author":"EA Paix\u00e3o","year":"2022","journal-title":"Cancers (Basel)"},{"issue":"3","key":"pcbi.1013312.ref040","doi-asserted-by":"crossref","first-page":"716","DOI":"10.1002\/cpt.2040","article-title":"Model-Based Cellular Kinetic Analysis of Chimeric Antigen Receptor-T Cells in Humans","volume":"109","author":"C Liu","year":"2021","journal-title":"Clin Pharmacol Ther"},{"key":"pcbi.1013312.ref041","unstructured":"Wu L, Trippe BL, Naesseth CA, Blei DM, Cunningham JP. Practical and Asymptotically Exact Conditional Sampling in Diffusion Models. In: 2024. https:\/\/arxiv.org\/abs\/2306.17775"}],"updated-by":[{"DOI":"10.1371\/journal.pcbi.1013312","type":"new_version","label":"New version","source":"publisher","updated":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T00:00:00Z","timestamp":1773964800000}}],"container-title":["PLOS Computational Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1013312","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T20:04:21Z","timestamp":1774037061000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1013312"}},"subtitle":[],"editor":[{"given":"Katharina","family":"Kusejko","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"editor"}]}],"short-title":[],"issued":{"date-parts":[[2026,3,16]]},"references-count":41,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3,16]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pcbi.1013312","relation":{},"ISSN":["1553-7358"],"issn-type":[{"value":"1553-7358","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,16]]}}}