Steps to reproduce analysis and figures

• Run make.R to prepare data for model fitting
• Run Rscripts/model_fitting/dothescience.job on HPC to fit model with multiple random starts
• Run additional scripts to generate additional posterior output
  • Targets:
    • cee_predictive_samples (to combine samples from multiple chains)
    • parameter_interpretation_patterns (interactive: to determine posterior predictive simulation starts)
• Run Rscripts/cee_prediction/dothescience.job on HPC to evaluate CEE responses
• Run Rscripts/parameter_interpretation/dothescience.job on HPC to evaluate covariate effects
• Run Rscripts/parameter_interpretation_ranefs/dothescience.job on HPC to evaluate random effects for each animal
• Run additional scripts and targets to generate figures and tables
  • Targets:
    • random_starts_convergence (Table 1; Supplement Figs. 1, 2)
    • sattag_illustration (Fig. 1)
    • parameter_interpretation_plot (Figs. 3, 4)
    • parameter_interpretation_variability_plot (Fig. 5)
    • cee_results (Figs. 6, 7)

Figures and tables are saved to disk under output/figures. Recall that Fig. 2 is a tikz image, built in the manuscript directly.

The scripts are not all done in a single workflow because parallelism is useful, and we deal with this differently. Note that we can change the script make.R so that it automatically kicks off the additional scripts to save some time, but, of course, we won't have the final scripts be able to be automatically run. Maybe, though, this is fine b/c we want to work on automating the generation of main model output, and we can deal with all of the additional post-processing separately.