Guide to serosolver terminology

James Hay

2025-06-12

Key terminology

The names below are used throughout the package documentation. They describe the main inputs, model quantities, and outputs rather than particular datasets.

Name	Meaning
individual	Identifier for a person. IDs should normally run from 1 to N without gaps.
sample_time	Time at which an antibody measurement was taken.
birth	Birth time used to determine which exposure times an individual could have experienced.
possible_exposure_times	Times at which an infection or other exposure can occur in the model.
biomarker_id	Identifier for the antigen, strain, or other biomarker being measured.
biomarker_group	Identifier for a separate observation and antibody-kinetics model, such as titre or avidity.
population_group	Group used to apply different infection-history priors or summarise attack rates.
measurement	Observed antibody or other biomarker value.
antibody_data	Data frame containing the serological measurements and their identifiers and times.
demographics	Optional data frame containing fixed or time-varying demographic covariates.
par_tab	Model control table containing parameter values, bounds, fixed/estimated status, and related settings.
inf_chain	MCMC draws for the latent infection histories.
theta_chain	MCMC draws for antibody-kinetics, observation, and related model parameters.

Current and published terminology

The current interface uses the following names in place of the names used by the published version of serosolver. The main change is from terminology based on influenza titres and strains to terminology that also covers other pathogens, assays, and biomarker types. See the published package and model description for the original notation and workflow.

Current name	Published name	Meaning
antibody_data	titre_dat	Main long-format serological data frame.
sample_time	samples	Time at which a sample was taken.
biomarker_id	virus	Identifier for the antigen, strain, or biomarker being measured.
measurement	titre	Observed antibody or biomarker value.
repeat_number	run	Identifier for a repeated measurement of the same sample and biomarker.
birth	DOB	Birth time used when determining possible exposure times.
population_group	group	Group used for infection-history priors and attack-rate summaries.
possible_exposure_times	strain_isolation_times or circulation_times	Times at which an infection or exposure can occur.
prior_version	version	Infection-history prior version used by the model.

biomarker_group and the separate demographics data frame are part of the expanded current interface and do not have a single direct equivalent in the published input table. They allow different observation types and fixed or time-varying covariates to be represented explicitly.

The figure below is from the serosim paper. It gives the corresponding mathematical notation and includes some additional model quantities. It is useful when comparing the current package interface with the notation used in the published version of serosolver, where some names and definitions differ.

Key serosolver variable terminology and its relationship to the model notation.