Fig. 3: Cancer cell fractions (CCFs) of subclones and mutations for patients P6, P4 and P1.

We computed the CCF for each mutation found in the tumor samples from patients P6, P4, and P1. Hierarchical clustering of mutations is then performed to obtain putative subclones. a, c, e Subclone CCF density figure shown as violin plots. The x-axis represents subclone clusters, and the n value shows the number of mutations in each subclone. Each mutation has two CCF values (as they may be detected in both tumor samples). The y-axis represents the density of CCFs for each subclone. b, d, f Scatter plot of CCFs for each patient. Each point represents a mutation. The x- and y-axis represent the CCFs of a mutation for each of two tumor samples in a patient. Mutations belonging to clusters n ≥ 3 are shown in their corresponding cluster colors in the violin plot. Mutations belonging to clusters n < 3 are outliers (shown in light grey) and not clustered. a 19 PyClone inferred subclone clusters in P6. b CCF relationship for P6. Subclone 1, 3, 4, 5, 9, and 11 in a are marked. PTEN, FBXW7, EIF4G3, IL12RB2, and PDE4B mutations are labelled with arrows. c Five PyClone inferred subclone clusters in P4. d CCF relationship for P4. Subclone 0, 1, and 2 in c are marked. e Seven PyClone inferred subclone clusters in P1. f CCF relationship for P1. Subclone 0, 1, 2, 3, 4, and 5 in e are marked. BCL11A mutation is labelled with arrows.