priors.R

prior_params = list(alpha_0_prior_mean = 5,
                    alpha_0_prior_sd = 2,
                    beta_0_prior_mean = -.5,
                    beta_0_prior_sd = 1,
                    trt_effect_sd = .5,
                    sigma_logvl_mean = 1,
                    sigma_logvl_sd = 1,
                    gamma_rnasep_sd = 1,
                    slope_coefs_sd = .5,
                    intercept_coefs_sd = .5,
                    t_max_pop_mean = -3,
                    t_max_pop_sd = 3)

prior_params_sensitive = prior_params
for(i in grep('sd',x = names(prior_params))){
  prior_params_sensitive[[i]] = prior_params_sensitive[[i]]*10
}

## we get convergence issues with the covariate models if you have very wide priors on the slope coefs
prior_params_sensitive$slope_coefs_sd=1

all_priors = list(WIP=prior_params, NIP=prior_params_sensitive)