Rxivist logo

Serosolver: an open source tool to infer epidemiological and immunological dynamics from serological data

By James A Hay, Amanda Minter, Kylie E. C. Ainslie, Justin Lessler, Adam J Kucharski, Steven Riley

Posted 08 Aug 2019
bioRxiv DOI: 10.1101/730069 (published DOI: 10.1371/journal.pcbi.1007840)

We present a flexible, open source R package designed to obtain additional biological and epidemiological insights from commonly available serological datasets. Analysis of serological responses against pathogens with multiple strains such as influenza pose a specific statistical challenge because observed antibody responses measured in serological assays depend both on unobserved prior infections and the resulting cross-reactive antibody dynamics that these infections generate. We provide a general modelling framework to jointly infer these two typically confounded biological processes using antibody titres against current and historical strains. We do this by linking latent infection dynamics with a mechanistic model of antibody dynamics that generates expected antibody titres over time. This makes it possible to use observations of antibodies in serological assays to infer an individual's infection history as well as the parameters of the antibody process model. Our aim is to provide a flexible inference package that can be applied to a range of datasets studying different viruses over different timescales. We present two case studies to illustrate how our model can infer key immunological parameters, such as antibody titre boosting, waning and cross-reaction, and well as latent epidemiological processes such as attack rates and age-stratified infection risk.

Download data

  • Downloaded 829 times
  • Download rankings, all-time:
    • Site-wide: 34,359
    • In immunology: 1,104
  • Year to date:
    • Site-wide: 139,974
  • Since beginning of last month:
    • Site-wide: 134,435

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide