Skip to main content
Figure 1 | BMC Bioinformatics

Figure 1

From: SubPatCNV: approximate subspace pattern mining for mapping copy-number variations

Figure 1

SubPatCNV workflow on a toy dataset. (a) CNV data are discretized. (b) The approximate pattern mining algorithm (SubPatCNV) is applied to discover frequent CNV patterns with the support threshold 0.5, error tolerant rate ε=0.2, and merge threshold δ=0.4. The itemsets in red are pruned by the support threshold or the merging criteria. The green itemsets are the maximum frequent itemsets by the criteria. Note that by convention approximate pattern mining does not allow any error tolerance for singleton itemsets and thus the itemset { i 1} is pruned. (c) CNV subspace patterns are visualized as the original log-intensity profiles.

Back to article page