{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T06:17:56Z","timestamp":1772173076458,"version":"3.50.1"},"reference-count":28,"publisher":"Public Library of Science (PLoS)","issue":"9","license":[{"start":{"date-parts":[[2022,9,8]],"date-time":"2022-09-08T00:00:00Z","timestamp":1662595200000},"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>Measles is one the best-documented and most-mechanistically-studied non-linear infectious disease dynamical systems. However, systematic investigation into the comparative performance of traditional mechanistic models and machine learning approaches in forecasting the transmission dynamics of this pathogen are still rare. Here, we compare one of the most widely used semi-mechanistic models for measles (TSIR) with a commonly used machine learning approach (LASSO), comparing performance and limits in predicting short to long term outbreak trajectories and seasonality for both regular and less regular measles outbreaks in England and Wales (E&amp;W) and the United States. First, our results indicate that the proposed LASSO model can efficiently use data from multiple major cities and achieve similar short-to-medium term forecasting performance to semi-mechanistic models for E&amp;W epidemics. Second, interestingly, the LASSO model also captures annual to biennial bifurcation of measles epidemics in E&amp;W caused by susceptible response to the late 1940s baby boom. LASSO may also outperform TSIR for predicting less-regular dynamics such as those observed in major cities in US between 1932\u201345. Although both approaches capture short-term forecasts, accuracy suffers for both methods as we attempt longer-term predictions in highly irregular, post-vaccination outbreaks in E&amp;W. Finally, we illustrate that the LASSO model can both qualitatively and quantitatively reconstruct mechanistic assumptions, notably susceptible dynamics, in the TSIR model. Our results characterize the limits of predictability of infectious disease dynamics for strongly immunizing pathogens with both mechanistic and machine learning models, and identify connections between these two approaches.<\/jats:p>","DOI":"10.1371\/journal.pcbi.1010251","type":"journal-article","created":{"date-parts":[[2022,9,8]],"date-time":"2022-09-08T13:31:12Z","timestamp":1662643872000},"page":"e1010251","update-policy":"https:\/\/doi.org\/10.1371\/journal.pcbi.corrections_policy","source":"Crossref","is-referenced-by-count":4,"title":["Comparing and linking machine learning and semi-mechanistic models for the predictability of endemic measles dynamics"],"prefix":"10.1371","volume":"18","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6590-0294","authenticated-orcid":true,"given":"Max S. Y.","family":"Lau","sequence":"first","affiliation":[]},{"given":"Alex","family":"Becker","sequence":"additional","affiliation":[]},{"given":"Wyatt","family":"Madden","sequence":"additional","affiliation":[]},{"given":"Lance A.","family":"Waller","sequence":"additional","affiliation":[]},{"given":"C. Jessica E.","family":"Metcalf","sequence":"additional","affiliation":[]},{"given":"Bryan T.","family":"Grenfell","sequence":"additional","affiliation":[]}],"member":"340","published-online":{"date-parts":[[2022,9,8]]},"reference":[{"key":"pcbi.1010251.ref001","doi-asserted-by":"crossref","first-page":"103791","DOI":"10.1016\/j.jbi.2021.103791","article-title":"Comparative study of machine learning methods for COVID-19 transmission forecasting","volume":"118","author":"A Dairi","year":"2021","journal-title":"Journal of Biomedical Informatics"},{"issue":"10","key":"pcbi.1010251.ref002","doi-asserted-by":"crossref","first-page":"e0005973","DOI":"10.1371\/journal.pntd.0005973","article-title":"Developing a dengue forecast model using machine learning: A case study in China.","volume":"11","author":"P Guo","year":"2017","journal-title":"PLOS Neglected Tropical Diseases"},{"issue":"1","key":"pcbi.1010251.ref003","doi-asserted-by":"crossref","first-page":"4413","DOI":"10.1038\/s41598-021-83926-2","article-title":"Infectious disease outbreak prediction using media articles with machine learning models.","volume":"11","author":"J Kim","year":"2021","journal-title":"Sci Rep"},{"key":"pcbi.1010251.ref004","article-title":"Time series analysis of COVID-19 infection curve: A change-point perspective.","author":"F Jiang","year":"2020","journal-title":"J Econom"},{"issue":"1","key":"pcbi.1010251.ref005","doi-asserted-by":"crossref","first-page":"129","DOI":"10.1186\/s12916-018-1108-5","article-title":"Neighbourhood level real-time forecasting of dengue cases in tropical urban Singapore.","volume":"16","author":"Y Chen","year":"2018","journal-title":"BMC Medicine"},{"issue":"5","key":"pcbi.1010251.ref006","doi-asserted-by":"crossref","first-page":"425","DOI":"10.1086\/286130","article-title":"Noise and Nonlinearity in Measles Epidemics: Combining Mechanistic and Statistical Approaches to Population Modeling.","volume":"151","author":"SP Ellner","year":"1998","journal-title":"The American Naturalist"},{"issue":"4968","key":"pcbi.1010251.ref007","doi-asserted-by":"crossref","first-page":"499","DOI":"10.1126\/science.2382131","article-title":"Chaos versus noisy periodicity: alternative hypotheses for childhood epidemics","volume":"249","author":"LF Olsen","year":"1990","journal-title":"Science"},{"issue":"2","key":"pcbi.1010251.ref008","doi-asserted-by":"crossref","first-page":"185","DOI":"10.1890\/0012-9615(2002)072[0185:DOMESN]2.0.CO;2","article-title":"Dynamics of Measles Epidemics: Scaling Noise, Determinism, and Predictability with the Tsir Model.","volume":"72","author":"BT Grenfell","year":"2002","journal-title":"Ecological Monographs"},{"issue":"9","key":"pcbi.1010251.ref009","doi-asserted-by":"crossref","first-page":"e1007305","DOI":"10.1371\/journal.pcbi.1007305","article-title":"Long-term dynamics of measles in London: Titrating the impact of wars, the 1918 pandemic, and vaccination.","volume":"15","author":"AD Becker","year":"2019","journal-title":"PLOS Computational Biology"},{"issue":"7179","key":"pcbi.1010251.ref010","doi-asserted-by":"crossref","first-page":"679","DOI":"10.1038\/nature06509","article-title":"The dynamics of measles in sub-Saharan Africa","volume":"451","author":"MJ Ferrari","year":"2008","journal-title":"Nature"},{"issue":"2","key":"pcbi.1010251.ref011","doi-asserted-by":"crossref","first-page":"e1004655","DOI":"10.1371\/journal.pcbi.1004655","article-title":"Persistent Chaos of Measles Epidemics in the Prevaccination United States Caused by a Small Change in Seasonal Transmission Patterns.","volume":"12","author":"BD Dalziel","year":"2016","journal-title":"PLOS Computational Biology"},{"issue":"49","key":"pcbi.1010251.ref012","doi-asserted-by":"crossref","first-page":"18438","DOI":"10.1073\/pnas.0603181103","article-title":"Inference for nonlinear dynamical systems","volume":"103","author":"EL Ionides","year":"2006","journal-title":"Proceedings of the National Academy of Sciences"},{"key":"pcbi.1010251.ref013","article-title":"Statistical Inference for Partially Observed Markov Processes via the R Package pomp.","author":"AA King","year":"2016","journal-title":"J Stat Soft"},{"issue":"1","key":"pcbi.1010251.ref014","doi-asserted-by":"crossref","first-page":"7061","DOI":"10.1038\/s41598-020-63877-w","article-title":"Algorithmic discovery of dynamic models from infectious disease data.","volume":"10","author":"J Horrocks","year":"2020","journal-title":"Sci Rep"},{"issue":"2204","key":"pcbi.1010251.ref015","doi-asserted-by":"crossref","first-page":"20170009","DOI":"10.1098\/rspa.2017.0009","article-title":"Model selection for dynamical systems via sparse regression and information criteria.","volume":"473","author":"NM Mangan","year":"2017","journal-title":"Proceedings of the Royal Society A: Mathematical, Physical and Engineering Sciences."},{"issue":"15","key":"pcbi.1010251.ref016","doi-asserted-by":"crossref","first-page":"3932","DOI":"10.1073\/pnas.1517384113","article-title":"Discovering governing equations from data by sparse identification of nonlinear dynamical systems.","volume":"113","author":"SL Brunton","year":"2016","journal-title":"Proceedings of the National Academy of Sciences"},{"issue":"1","key":"pcbi.1010251.ref017","doi-asserted-by":"crossref","first-page":"563","DOI":"10.1186\/s12889-016-3239-y","article-title":"Factors associated with performing tuberculosis screening of HIV-positive patients in Ghana: LASSO-based predictor selection in a large public health data set.","volume":"16","author":"S Mueller-Using","year":"2016","journal-title":"BMC Public Health"},{"issue":"5","key":"pcbi.1010251.ref018","doi-asserted-by":"crossref","first-page":"e1007839","DOI":"10.1371\/journal.pcbi.1007839","article-title":"Inference on dengue epidemics with Bayesian regime switching models.","volume":"16","author":"JT Lim","year":"2020","journal-title":"PLoS Comput Biol"},{"key":"pcbi.1010251.ref019","article-title":"A competing-risks model explains hierarchical spatial coupling of measles epidemics en route to national elimination","author":"MSY Lau","year":"2020","journal-title":"Nature Ecology & Evolution"},{"issue":"1","key":"pcbi.1010251.ref020","doi-asserted-by":"crossref","first-page":"48","DOI":"10.2307\/2342553","article-title":"Measles Periodicity and Community Size","volume":"120","author":"MS Bartlett","year":"1957","journal-title":"Journal of the Royal Statistical Society Series A"},{"key":"pcbi.1010251.ref021","doi-asserted-by":"crossref","first-page":"462","DOI":"10.1007\/1-4020-0612-8","volume-title":"Encyclopedia of Production and Manufacturing Management","author":"PM Swamidass","year":"2000"},{"issue":"12","key":"pcbi.1010251.ref022","doi-asserted-by":"crossref","first-page":"121102","DOI":"10.1063\/1.5010300","article-title":"Using Machine Learning to Replicate Chaotic Attractors and Calculate Lyapunov Exponents from Data.","volume":"27","author":"J Pathak","year":"2017","journal-title":"Chaos"},{"issue":"2","key":"pcbi.1010251.ref023","doi-asserted-by":"crossref","first-page":"024102","DOI":"10.1103\/PhysRevLett.120.024102","article-title":"Model-Free Prediction of Large Spatiotemporally Chaotic Systems from Data: A Reservoir Computing Approach","volume":"120","author":"J Pathak","year":"2018","journal-title":"Phys Rev Lett"},{"issue":"4","key":"pcbi.1010251.ref024","doi-asserted-by":"crossref","first-page":"041101","DOI":"10.1063\/1.5028373","article-title":"Hybrid Forecasting of Chaotic Processes: Using Machine Learning in Conjunction with a Knowledge-Based Model.","volume":"28","author":"J Pathak","year":"2018","journal-title":"Chaos"},{"issue":"2","key":"pcbi.1010251.ref025","first-page":"187","article-title":"Time series modelling of childhood diseases: a dynamical systems approach","volume":"49","author":"BF Finkenst\u00e4dt","year":"2000","journal-title":"Journal of the Royal Statistical Society: Series C (Applied Statistics)"},{"issue":"2","key":"pcbi.1010251.ref026","doi-asserted-by":"crossref","first-page":"169","DOI":"10.1890\/0012-9615(2002)072[0169:DOMEES]2.0.CO;2","article-title":"Dynamics of Measles Epidemics: Estimating Scaling of Transmission Rates Using a Time Series Sir Model","volume":"72","author":"ON Bj\u00f8rnstad","year":"2002","journal-title":"Ecological Monographs"},{"issue":"9","key":"pcbi.1010251.ref027","doi-asserted-by":"crossref","first-page":"e0185528","DOI":"10.1371\/journal.pone.0185528","article-title":"tsiR: An R package for time-series Susceptible-Infected-Recovered models of epidemics.","volume":"12","author":"AD Becker","year":"2017","journal-title":"PLOS ONE."},{"issue":"1","key":"pcbi.1010251.ref028","doi-asserted-by":"crossref","first-page":"267","DOI":"10.1111\/j.2517-6161.1996.tb02080.x","article-title":"Regression Shrinkage and Selection via the Lasso","volume":"58","author":"R. Tibshirani","year":"1996","journal-title":"Journal of the Royal Statistical Society Series B (Methodological)"}],"container-title":["PLOS Computational Biology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1010251","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,9,8]],"date-time":"2022-09-08T13:31:53Z","timestamp":1662643913000},"score":1,"resource":{"primary":{"URL":"https:\/\/dx.plos.org\/10.1371\/journal.pcbi.1010251"}},"subtitle":[],"editor":[{"given":"T. Alex","family":"Perkins","sequence":"first","affiliation":[]}],"short-title":[],"issued":{"date-parts":[[2022,9,8]]},"references-count":28,"journal-issue":{"issue":"9","published-online":{"date-parts":[[2022,9,8]]}},"URL":"https:\/\/doi.org\/10.1371\/journal.pcbi.1010251","relation":{"has-preprint":[{"id-type":"doi","id":"10.1101\/2022.05.27.22275695","asserted-by":"object"}]},"ISSN":["1553-7358"],"issn-type":[{"value":"1553-7358","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,9,8]]}}}