Figure 7From: Parallel Markov chain Monte Carlo - bridging the gap to high-performance Bayesian computation in animal breeding and geneticsHistogram of posterior estimates of 1- π in the BayesC π model for feature selection. The results were obtained from feature selection using computing strategy II, that is, one round of parallel FS jobs was then executed, followed by 30 rounds of parallel post-FS inference and CV jobs conducted sequentially for all panel sizes.Back to article page