A Bayesian modelling framework for estimating tick-borne pathogen transmission dynamics at the host-tick interface

Journal article

Understanding the transmission dynamics of tick-borne pathogens at the host-tick interface is challenged by the presence of multiple pathways for tick infection, including (i) host-to-tick transmission, (ii) tick-to-tick (cofeeding) transmission, and (iii) pre-existing infection through vertical transmission or prior feeding. Assessing parameters governing these pathways is critical for identifying the main transmission drivers and, consequently, key prevention and control points. Here, we developed a Bayesian modelling framework that estimates key parameters describing the probability of each transmission pathway and assesses associated factors, including bird species, tick life stage and engorgement level, by jointly modelling transmission at the host-tick interface using data collected in field studies that sample hosts and their ticks. First, by fitting the model to simulated host-tick infection data, we demonstrated the framework’s ability to recover the parameter values underlying these data. Model performance improved significantly when more information was available on variability in cofeeding probability among individual ticks, highlighting the value of testing all collected ticks and recording their spatial distribution on the host in relation to each other. Second, we fitted the model to field data collected at the bird-tick interface in Northeast France in 2023, focusing on Borrelia garinii, B. valaisiana, and Anaplasma phagocytophilum as case pathogens. For all three pathogens studied, models including cofeeding transmission explained the data significantly better than models that did not. Engorgement level was significantly and positively associated with the probability of bird-to-tick transmission for A. phagocytophilum. Finally, the estimated parameters, such as the probability of A. phagocytophilum infection in birds and the probability of Borrelia or Anaplasma infection in ticks before feeding, were comparable to values from an external dataset, not used for model fitting. Our framework provides a valuable foundation for future research to understand tick-borne pathogen transmission dynamics based on epidemiological and ecological field data collected at the host-tick interface.

Authors: Kim, Y; Faivre, B; Boulinier, T; Sineau, C; Galon, C

• Publication status: Published
• Peer review status: Peer reviewed
• Publisher: Public Library of Science
• Journal: PLoS Computational Biology
• Volume: 22
• Issue: 4
• Pages: e1014146
• Article number: e1014146
• Publication date: 2026-04-03
• Acceptance date: 2026-03-20
• DOI: 10.1371/journal.pcbi.1014146
• EISSN: 1553-7358
• ISSN: 1553734X, 1553-734X
• Language: English
• Keywords: Transmission (telecommunications); Interface (matter); Bayesian probability; Bayes' theorem; Field (mathematics); Bayesian inference; Probability distribution; Key (lock)