Example mechanistic trajectory model: eclipse-phase viral kinetics

A target-cell-limited within-host viral kinetic model with an eclipse phase (Eq. 11-14 of Do, McCarthy & Rogers / Baccam et al. 2006), provided as a ready-to-use example of the simulate_fn ingredient MBMM needs (see mbmm_abc_fit). Integration uses deSolve (lsoda), because this system is numerically stiff enough that a naive fixed-step integrator (like mbmm_rk4_solve) can become unstable at the eclipse/clearance rates used here; deSolve is a package dependency for exactly this reason. Fixed constants (k, c) and prior locations below are illustrative, not pathogen-specific – re-derive them from the literature for a real analysis (as Sec. 4.2/App. A.1 of the manuscript do from UKC19 human challenge data).

Usage

mbmm_example_eclipse_simulate(theta_row, times, T0 = 1e+07, k = 4, c = 10)

Arguments

theta_row

a one-row data.frame/list with elements beta, delta, p, V0, offset

times

observed times (e.g. "day post enrollment"), any non-negative numeric vector

T0

fixed initial target-cell count (default 1e7)

k

fixed eclipse-to-productive transition rate, 1/day (default 4)

c

fixed free-virus clearance rate, 1/day (default 10)

Value

numeric vector of predicted viral load (natural scale, e.g. copies/mL) at each element of times

Details

Free parameters (columns expected in theta_row): beta (infection rate), delta (clearance rate of infected cells), p (viral production rate), V0 (initial viral burden), and offset (days between the model's internal time origin and the cohort's observed day-0, allowing unobserved infection timing to vary across subjects – see Sec. 4.1 of the manuscript).