WEHI PhD Completion Seminar hosted by Professor Ivo Mueller

Felicia Bongiovanni

PhD Student – Mueller Laboratory, Immunity division, WEHI

When did Bob get sick?

Mathematical modelling of antibody kinetics for the inference of time-since-infection

Davis Auditorium

Join via TEAMS

Including Q&A session

Can you remember the last time you were sick, down to the exact day? Knowing when an individual was infected can help diagnose disease and understand how long protective immunity lasts. At the population level, knowing infection timing can inform key epidemiological measures such as incidence and the force of infection. The literature includes multiple methods to estimate time-since-infection (TSI) from serological data, with machine learning approaches such as random forest classifiers being prominent. Although powerful, random forest classifiers return a binary or categorical classification rather than a continuous estimate. Other methods involving statistical frameworks also exist and can provide detail on immunological and epidemiological dynamics; however, the design and complexity of these approaches limit their accessibility and widespread use. 

In this PhD completion seminar, I present a hierarchical Bayesian framework that combines a mechanistic antibody kinetics model with serological data to characterise humoral immune dynamics and estimate TSI on a continuous time scale. Validated across multiple diseases and implemented as an open-source R package to make it accessible to other researchers, the framework is illustrated through the case of Bob, a hypothetical individual, to show how it can: a) reconstruct antibody response curves and b) use information on antibody dynamics to estimate TSI, bringing us back to the central question: when did Bob get sick?

All welcome!