Fig 1.
PanSim maps simulated population movements and projected infections onto specific locations in the virtual city emulating Szeged.
(A) Institutions and households are localised in a virtual city. Colour codes represent different types of locations, and node sizes increase linearly with the number of infected people (see an animated version of this map at https://youtu.be/OCfbHjLeCbY). (B) Distribution of infection events at various types of locations when the virus is unmitigated or when restrictions from the reference scenario are applied. (Note the logarithmic scale, error bars show the uncertainty of 20 simulations.) (C) Projected sensitivity of the pandemic with various intervention types and levels. Colour code shows the severity function of the pandemics for the labelled changes. (Details on the analysed scenarios can be found in Table F in S1 Text. (D) Comparing actual hospitalisation data [36] to the fitted and the reference scenario simulations. (Mean and std. of 30 simulations are shown).
Fig 2.
Variations in quarantine scenarios and testing intensities can slow down but not suppress the pandemics.
(A-B) Time course of simulations with varied quarantine scenarios (Q1—diagnosed patient quarantined at home, Q2—diagnosed patient and household quarantined, Q3—patient, household and workmates/classmates are quarantined) showing (A) the percentage of hospitalised COVID-19 patients in the population and (B) the accumulated number of death events due to COVID-19 scaled to the whole population. Appearance and spread of the B.1.1.7 variant are shown on the inset of panel A. Start date of restrictions and the start of vaccination are noted on plots. (C-D) The testing rate was increased from the reference scenario value of 0.15% of people tested daily, fitted to Hungarian data (Fig 1D and Fig E in S1 Text) to the highest level achieved by most actively testing countries (3.5% daily). (C) Time courses of the daily ratio of positive tests and (D) the percentage of hospitalised COVID-19 patients in the population each day (hospital burden) are plotted. Daily testing rates are shown on the inset of panel C. (Mean and std. of 30 simulations are shown).
Fig 3.
Results of various vaccination and reopening strategies.
(A) Percentage of the population infected, (B) percentage of hospitalised COVID-19 patients in the population, (C) the accumulated number of death events due to COVID-19 scaled to the whole population each day. Panel A-C were simulated with the assumption that daily, 0.2% of the population could be vaccinated. (D) Changes in this daily vaccination rate have major effects on the percentage of hospital patients. (Mean and std. of 30 simulations are shown).
Fig 4.
High infection rate of children and precise, effective reproduction number (Rt) emerge from the simulation results.
(A) Percentage of the various age groups catching the infection during the autumn and the spring wave (error bars show the uncertainty of 20 simulations). (B) Changes in the reproduction number of the virus were calculated from the fitted scenario (Fig 1D) simulations (mean and std. of 30 simulations) plotted together with the empirical data of Hungary [22].