Configs-class: Class 'Configs'
In wlandau/fbseq: fbseq

Description Slots

Set individual MCMC control parameters (slots listed in help("Configs-class")).

burnin: MCMC burnin, the number of MCMC iterations to ignore at the beginning of each obj
effects_update_beta: values of l for which to update the beta_l, g parameters. For debugging only.
theta_update: Indices l for which theta_l is updated/sampled in the MCMC.
genes_return: Indices of genes whose parameter samples you want to return. Applies to all gene-specific parameters except for the epsilons.
genes_return_epsilon: Indices of genes g for which epsilon_n, g is updated/returned.
iterations: Number of MCMC iterations after burnin for which selected parameter samples are kept. Total MCMC iterations = burnin + thin * "iterations", and the whole "thin * iterations" portion is used to calculate posterior means, mean squares, and probabilities.
libraries_return: Indices of RNA-seq libraries whose parameter samples you want to return. Currently moot because there are no library-specific parameters other than the epsilons, but that could change in future versions of the package.
libraries_return_epsilon: Indices of RNA-seq libraries n for which epsilon_n, g is updated/returned. Applies to all library-specific parameters except for the epsilons.
parameter_sets_return: Character vector naming the variables whose MCMC samples you want to return
parameter_sets_update: Character vector naming the variables to calculate/update during the MCMC.
priors: Character vector. Names of the family of priors on the betas after integrating out the xi's. Can be any value returned by special_beta_priors(). All other values will default to the normal prior.
samplers: character string indicating the sampling algorithm
thin: MCMC thinning interval. thin = 1 means parameter samples will be saved for every iterations after burnin. thin = 10 means parameter samples will be saved every 10th iteration after burnin. Total MCMC iterations = burnin + thin * "iterations", and the whole "thin * iterations" portion is used to calculate posterior means, mean squares, and probabilities.
verbose: Number of times to print out progress during burnin and the actual MCMC. If verbose > 0, then progress messages will also print during setup and cleanup.