COVID-19 Enhanced Shielding Model

Repository for R and C++ code used to model enhanced shielding measures on a simulated COVID-19 outbreak.

Running the Code

R

R code should run without further modification (after installing pre-requisite packages). The chosen directory where plots are saved can be altered by changing the setwd() function at the top of each script. Plots are automatically saved in the chosen working directory if each script is run.

Code to run analysis of the population structure Enhanced_Shielding_PopStruct_FigS3_5.R (Figure S3 & 5) require the corresponding .csv files to be present in the User's chosen working directory. The name of .csv correspond to each population structure analysed and represent simulations implemented in C++ and imported into R for graphical output.

C++

C++ code can be found in main.cpp. This file can be compiled using g++. Implementation makes use of boost libraries (odeint).

Navigating the Repository

R

R code can be found in the Enhanced Shielding folder and is organised according to the code used to plot figures found in the manuscript. WORK IN PROGRESS Current code is available for:

• Figure 2 & S1 - Baseline Analysis using "Central" Parameters and Recovered Fraction Plot
  • Enhanced_Shielding_Baseline_Fig2_S1.R
• Figure 3 - Sensitivity Analysis for Phase4 R_e, Zeta (1/Duration of Immunity), Phase 1 R_e and % Compliance of Vulnerable Population
  • Enhanced_Shielding_Phase4_Zeta_Phase1_Compliance_Fig3.R
• Figure S2 - Sensitivity analysis of Phase 2 lockdown duration
  • Enhanced_Shielding_LockdownDur_S2.R
• Figure S3 - Sensitivity Analysis for Phase3 R_e Duration
  • Enhanced_Shielding_Phase3_FigS3.R
• Figure S4 & S5 - Sensitivity analysis of the lockdown trigger day and the Phase 2 R_e
  • Enhanced_Shielding_TrigDay_Phase2_FigS4_S5.R
• Figure S6 & 7 - Sensitivity analysis differing population structure and vulnerable/shielders ratio
  • Enhanced_Shielding_PopStruct_FigS6_7.R
• Figure S8 - SIS Baseline Plot
  • Enhanced_Shielding_SIS_FigS8.R
• Figure S9 - Analysis of efficacy of Shielders % testing
  • Enhanced_Shielding_ShieldTest_FigS9.R

Programs and Packages Used

COVID-19 modelling code was implemented using R (3.6.2) and C++ independently. ODEs were solved using the desolve (1.27.1) package in R and odeint in C++. Plotting in R was carried out using the ggplot package (3.3.0). Dataframe manipulation was performed using reshape2 (1.4.4). Finalised plot output was performed using ggpubr (0.2.4) and Cairo (1.5-10) packages.

Acknowledgements

All analysis were performed by members and affiliates of Epigroup, University of Edinburgh: https://www.wiki.ed.ac.uk/display/Epigroup/Epigroup+-+Home