Plots of predictive checks for mixture prior.
Histograms and q-q plots of predictive p-values for the mixture prior. Separate plots are given for each mixture component, using only genes with high posterior probability of being classified into the relevant component.
plotPredChecks(pvals, pc, probz = 0.8, label = "", breaks = 20)
- matrix of predictive p-values output by 'ccPred' (NB, not the whole list object, just the matrix of p-values)
- matrix of posterior classification probabilities output by 'ccParams' (NB, not the whole list object, just the matrix of probabilities)
- threshold on posterior probabilities for including genes in each mixture component plot
- title used on histograms
- argument input to histogram
## Note this is a very short MCMC run! ## For good analysis need proper burn-in period. data(ybar,ss) outdir <- BGmix(ybar, ss, c(8,8), nburn=0, niter=100, nthin=1) params <- ccParams(outdir) pred <- ccPred(outdir) plotPredChecks(pred$pval.ybar.mix2,params$pc,probz=0.5)
Looks like there are no examples yet.