The assign.summary function computes the posterior mean of the model parameters estimated in every iteration during the Gibbs sampling.
assign.summary(test, burn_in = 1000, iter = 2000, adaptive_B = TRUE, adaptive_S = FALSE, mixture_beta = TRUE)
The list object returned from the assign.mcmc function. The list components are the MCMC chains of the B, S, Delta, beta, gamma, and sigma.
The number of burn-in iterations. These iterations are discarded when computing the posterior means of the model parameters. The default is 1000.
The number of total iterations. The default is 2000.
Logicals. If TRUE, the model adapts the baseline/background (B) of genomic measures for the test samples. The default is TRUE.
Logicals. If TRUE, the model adapts the signatures (S) of genomic measures for the test samples. The default is FALSE.
Logicals. If TRUE, elements of the pathway activation matrix are modeled by a spike-and-slab mixture distribution. The default is TRUE.
The N x K matrix of the posterior mean of the pathway activation level in test samples (transposed matrix A). Columns:K pathways; rows: N test samples
The G x 1 vector of the posterior mean of the variance of gene.
The N x K matrix of posterior mean of pathway activation level in test samples (transposed matrix A) (adjusted beta_pos scaling between 0 and 1). Columns:K pathways; rows: N test samples
The N x K matrix of the posterior probability of pathways being activated in test samples.
The G x K matrix of the posterior mean of pathway signature genes.
The G x K matrix of the posterior probability of genes being significant in the associated pathways.
The assign.summary function is suggested to run after the assign.convergence function, which is used to check the convergence of the MCMC chain. If the MCMC chain does not converge to a stationary phase, more iterations are required in the assign.mcmc function. The number of burn-in iterations is usually set to be half of the number of total iterations, meaning that the first half of the MCMC chain is discarded when computing the posterior means.
data(trainingData1) data(testData1) data(geneList1) trainingLabel1 <- list(control = list(bcat=1:10, e2f3=1:10, myc=1:10, ras=1:10, src=1:10), bcat = 11:19, e2f3 = 20:28, myc= 29:38, ras = 39:48, src = 49:55) processed.data <- assign.preprocess(trainingData=trainingData1, testData=testData1, trainingLabel=trainingLabel1, geneList=geneList1)#>#>mcmc.chain <- assign.mcmc(Y=processed.data$testData_sub, Bg = processed.data$B_vector, X=processed.data$S_matrix, Delta_prior_p = processed.data$Pi_matrix, iter = 20, adaptive_B=TRUE, adaptive_S=FALSE, mixture_beta=TRUE)#>#>mcmc.pos.mean <- assign.summary(test=mcmc.chain, burn_in=10, iter=20, adaptive_B=TRUE, adaptive_S=FALSE, mixture_beta = TRUE)