Skip to main content
Figure 3 | BMC Bioinformatics

Figure 3

From: Detection of copy number variation from array intensity and sequencing read depth using a stepwise Bayesian model

Figure 3

Model selection. (A) A simulated four-CNV array-CGH data set. The data D (M = 1000) were generated with θ= {N = 4, (s1 = 100, w1 = 30, a1 = 0.7), (s2 = 200, w2 = 20, a2 = -0.3), (s3 = 400, w3 = 80, a3 = 1.5), (s4 = 600, w4 = 90, a4 = -0.6), σ2 = 0.12}. The scatter plot of the multi-CNV array-CGH data are overlaid with the segmentation found by our algorithm using different models. The step function lines in red, green, blue, and orange represent models with N = 1, 2, 3, and 4, respectively. For figure clarity, the estimated model with N = 5 is not shown. (B) Model comparison. The comparison between each of the models with N = 1, 2, 3, 4 and the basal, null model (N = 0) was quantified by the logarithm of the Bayes factor.

Back to article page