Dynamic causal modelling of immune heterogeneity

Abstract An interesting inference drawn by some COVID-19 epidemiological models is that there exists a proportion of the population who are not susceptible to infection—even at start current pandemic. This paper introduces model immune response virus. based upon same sort mean-field dynamics as used in epidemiology. However, place location, clinical status, and other attributes people an model, we consider state virus, B T-lymphocytes, antibodies they generate. Our aim formalise key hypotheses mechanism resistance. We present series simple simulations illustrating changes under these hypotheses. These include attenuated viral cell entry, pre-existing cross-reactive humoral (antibody-mediated) immunity, enhanced T-cell dependent immunity. Finally, illustrate potential application this variational inversion (using simulated data) its use testing In principle, furnishes fast efficient immunological assay—based on sequential serology—that provides (1) quantitative measure latent responses (2) Bayes optimal classification different kinds (c.f., glucose tolerance tests test for insulin resistance). may be especially useful assessing SARS-CoV-2 vaccines.