Figure 2

Simulation design and patterns discovered by iASeq. (a) The true ASB patterns in simulation 1. Two patterns were simulated in addition to the background pattern. The two non-background patterns are shown. Each pattern has 4725 SNPs. Each row in the plot represents a SNP class, and each column represents a dataset. Black means skewed, and white means not skewed. (b) The BIC values for different class number K in simulation 1. The BIC achieves the minimum at K=2. (c) Patterns discovered by iASeq in simulation 1. The plot shows the estimated V and W when K=2. Each row corresponds to a class. Each column represents a dataset. The color in the cell (k,d) demonstrates the estimated SR or SN probability in class k and dataset d. From white to dark, the probability increases from 0 to 1. The numbers shown under Π are the estimated number of SNPs in each class (i.e., ${\widehat{\Pi }}_k\ast$ the total number of SNPs). The numbers shown under a _i are the number of SNPs identified for the corresponding class using the posterior probability Pr(a _i =k|X _i ,N _i ,Π,V,W)>0. 9 as cutoff. (d) The true ASB patterns in simulation 2. Four patterns were simulated in addition to the background pattern. The four non-background patterns are shown. Each pattern has 2362 SNPs. (e) The BIC values for different class number K in simulation 2. The BIC achieves the minimum at K=4. (f) The patterns discovered by iASeq in simulation 2.