Stephen M. Kissler, James A. Hay, Joseph R. Fauver, Christina Mack, Caroline G. Tai, Deverick J. Anderson, David D. Ho, Nathan D. Grubaugh, Yonatan H. Grad
This repository accompanies the manuscript "Viral kinetics of sequential SARS-CoV-2 infections". The scripts (in code/
) are:
run_analysis.R
: This is the main runfile, calling all of the other scripts to generate the figures and tables presented in the manuscript.utils.R
: contains key functionsutils_private.R
: contains a single line formatted asextdrive <- "~/Local/Output/Directory/"
for storing large output filesset_global_pars.R
sets global parameters (currently just the qPCR limit of detection)set_run_pars.R
defines any exclusions, stratifications, and priors for the model fits. Each element of the list defined in this script defines a different analysis (i.e., a different set of comparison groups).fit_posteriors_preamble.R
formats the data in a way that's interpretable by Stan.fit_posteriors.R
calls the Stan script for fitting the model.fit_posteriors.stan
is the Stan script where the model is implemented and fit.make_figures.R
makes essential figures displayed in the manuscriptsave_figures.R
saves those figuresclearbigvars.R
clears storage-heavy variables to avoid crashing R or needlessly slowing down operations.indiv_analysis_revised.R
runs the Spearman correlation analysis to detect persistence in viral kinetic parameters within individuals across first and second infections.