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

Description Slots

MCMC Chain class.

bound_names: names of bounds slot in the original Scenario object
contrast_names: names of contrasts slot in the Scenario object
gene_names: names of genes, taken from the row names of the count data matrix
library_names: names of the libraries/samples, taken from the column names of the count data matrix
proposition_names: names of propositions slot in the Scenario object
bounds: values to compare contrasts to. The comparison is to see if the contrast is greater than its corresponding element in bounds (from the Scenario object)
contrasts: contrasts from the Scenario object.
counts: RNA-seq count data, flattened from a matrix
design: Design, flattened from the design matrix. Original matrix must have rows corresponding to colums/libraries in RNA-seq data and colums corresponding to sets of gene-specific variables.
propositions: propositions of inequalities involving contrasts from the Scenario object.
supplement: a list containing supplementary information about the scenario: for example, how the data were simulated, if applicable
burnin: MCMC burnin, the number of MCMC iterations to ignore at the beginning of each obj
effects_update_beta: bounds of l for which to update the beta_l, g parameters.
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: 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 bounds will default to the normal prior.
samplers: character string specifying 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.
C: number of contrasts
countSums_g: gene-specific count sums
countSums_n: library-specific count sums
designUnique: Matrix of unique nonzero elements of design. Vacent entries are 0.
designUniqueN: for each column index l, number of unique nonzero elements of design[, l].
G: number of genes
Greturn: number of genes to return gene-specific MCMC parameter samples for (except the epsilons)
GreturnEpsilon: number of genes to return gene-specific MCMC epsilon parameter samples
L: number of columns in the original design matrix
Lupdate_beta: number of bounds of l for which to update the beta_l, g parameters.
Lupdate_theta: number of indices l for which to update the theta_l parameters.
N: number of libraries
Nreturn: number of libraries to return library-specific MCMC parameter samples for (except the epsilons)
NreturnEpsilon: number of libraries to return library-specific MCMC epsilon parameter samples
P: number of propositions involving contrasts
probs: estimated posterior probabilities of propositions in the Scenario object.
contrastsPostMean: Posterior means of the linear combinations of the betas, specified by the contrasts slot.
contrastsPostMeanSquare: Posterior mean squares of the linear combinations of the betas, specified by the contrasts slot.
seeds: vector of N*G random number generator seeds
a: initialization constant
b: initialization constant
c: initialization constants
d: initialization constant
h: initialization constants
k: initialization constants
q: initialization constants
r: initialization constants
s: initialization constants
beta: MCMC parameter samples
epsilon: MCMC parameter samples
gamma: MCMC parameter samples
nu: MCMC parameter samples
sigmaSquared: MCMC parameter samples
tau: MCMC parameter samples
theta: MCMC parameter samples
xi: MCMC parameter samples
betaStart: MCMC starting bounds
epsilonStart: MCMC starting bounds
gammaStart: MCMC starting bounds
nuStart: MCMC starting bounds
sigmaSquaredStart: MCMC starting bounds
tauStart: MCMC starting bounds
thetaStart: MCMC starting bounds
xiStart: MCMC starting bounds
betaPostMean: estimated posterior means
epsilonPostMean: estimated posterior means
gammaPostMean: estimated posterior means
nuPostMean: estimated posterior mean
sigmaSquaredPostMean: estimated posterior means
tauPostMean: estimated posterior mean
thetaPostMean: estimated posterior means
xiPostMean: estimated posterior means
betaPostMeanSquare: estimated posterior means of the squares of parameters
epsilonPostMeanSquare: estimated posterior means of the squares of parameters
gammaPostMeanSquare: estimated posterior means of the squares of parameters
nuPostMeanSquare: estimated posterior mean of the square of the parameter
sigmaSquaredPostMeanSquare: estimated posterior means of the squares of parameters
tauPostMeanSquare: estimated posterior mean of the square of the parameter
thetaPostMeanSquare: estimated posterior means of the squares of parameters
xiPostMeanSquare: posterior means of the squares of parameters
betaSampler: sampler option
epsilonSampler: sampler option
gammaSampler: sampler option
nuSampler: sampler option
sigmaSquaredSampler: sampler option
tauSampler: sampler option
thetaSampler: sampler option
xiSampler: sampler option
betaTune: tuning parameter
epsilonTune: tuning parameter
gammaTune: tuning parameter
nuTune: tuning parameter
sigmaSquaredTune: tuning parameter
tauTune: tuning parameter
thetaTune: tuning parameter
xiTune: tuning parameter