Skip to main content
Figure 4 | Genome Biology

Figure 4

From: A general framework for analyzing tumor subclonality using SNP array and DNA sequencing data

Figure 4

Identifiability zones in sAGP-CCF space, for amplification (A), CN-LOH (B) and deletion (C), with up to four scenarios described in the main text. Theoretically permissible areas of sAGP-CCF for different scenarios are marked by borders of different colors, and labeled with a single letter (such as `A1’) for uniquely occupied zones, and by two or more letters (`A1/C’) for overlapping zones. Regions of light gray support a unique CCF expression, whereas the regions of dark gray cannot unambiguously estimate CCF. The density contours (in orange) depict the distribution of 3,382 mutations in amplification regions (A), 2,008 in CN-LOH, and 4,662 in deletions representing 201 breast tumor samples with least data loss in sAGP estimation. Variants with coverage lower than 20 or SAF smaller than 0.05 were discarded. Only a very small portion of the mutations fall outside the theoretically predicted zones. Among the rest, approximately 48% belong to a unique scenario, but approximately 93% have a unique CCF estimation.

Back to article page