analysis of toy_outbreak_long dataset from the o2geosocial package

install package o2geosocial from "alxsrobert/o2geosocial". To generate all figures from main and supplement, run the script generate_figure.R.

install.packages("devtools")
library(devtools)
install_github("alxsrobert/o2geosocial")
library(o2geosocial)
source("R/generate_figures.R", echo = TRUE)

To generate a new simulated dataset with different input parameters (number of cases, R0, spatial parameters...), edit the lines 8-40 in analysis_generated_data.R. Run the rest of the script to generate the o2geosocial runs on the dataset.

List of files in the repository:

• In data:

  • State_initials.csv : Data table with state names and acronyms.
  • pop_center.csv: Population centroid and population of every county (Source: https://www.census.gov/geographies/reference-files/2010/geo/2010-centers-population.html (per state : UNITED STATES)).

• In R:

  • library_importation.R: import libraries for analysis.
  • function_generate_dataset.R: Functions to generate toy_outbreak_long.
  • function_prepare_for_figures.R: Function to generate summary statistics on run.
  • function_generate_figures_main.R: function to generate figures similar to figure 3, 4, 5 and 6.
  • function_supplement_figures.R: function to generate figures similar to figures in the supplement.
  • load_all_analysis.R: Load all o2geosocial runs.
  • generate_figs_analysis_threshold.R: Load o2geosocial runs with different import thresholds and save files to generate figures.
  • generate_figs_analysis_likelihoods.R: Load o2geosocial runs with different likelihoods and save files to generate figures.
  • generate_figs_analysis_genotype.R: Load o2geosocial runs with different proportion of genotyped cases and save files to generate figures.
  • load_data_distributions.R: Load prior distributions.
  • analysis_generated_data.R: Script to run o2geosocial on toy_outbreak_long.
  • generate_figure.R: Generate all figures (main + supplement).