Fig. 2: Mitochondrial mutations serve as high-confidence clonal markers in AML. | Nature Communications

Fig. 2: Mitochondrial mutations serve as high-confidence clonal markers in AML.

From: Identification of leukemic and pre-leukemic stem cells by clonal tracking from single-cell transcriptomics

Fig. 2: Mitochondrial mutations serve as high-confidence clonal markers in AML.

See also Supplementary Figs. 26. a Overview of the patients. CD34+ and CD34- indicate the dominant surface phenotype of the leukemic blasts (see Fig. 3b and Supplementary Fig. 2 for quantification). For each patient, genes containing mutations are printed in italic (see Supplementary Data 1 for a complete list). Subcl., sub-clonal mutation. b Heatmap depicting variant allele frequencies (color coded, see right of Fig. 2c for a key) observed in single-cell RNA-seq data of n = 1430 cells from P1. Gray indicates missing values. Cells and mutations are arranged according to the clustering result obtained by PhISCS36 as described in the “Methods” section, Analysis of mitochondrial mutations and reconstruction of clonal hierarchies. Calculation of the likelihood is described in the same section. Mutations with low coverage (in TET2, and SPEN) were not included in the clustering and are depicted in the heatmaps as metadata, however, in all cases except TET2 frameshift there is quantitative evidence for their association with specific clones (Supplementary Fig. 5b). For reproducing the computations, see the vignettes accompanying the mitoClone package. For mutations, nuc is nuclear genome; mt is mitochondrial genome. c Like panel b, but using n = 1066 cells from P2. DNMT3A is included as a low coverage mutation with significant association to the pre-leukemic and leukemic clone (Supplementary Fig. 5b). d Heatmap depicting variant allele frequencies (color coded, see legend to the right of Fig. 2c) observed in targeted DNA-seq data from n = 288 single-cell derived colonies from P1. e PhISCS36 was run on the mutational data from P1 to reconstruct a clonal hierarchy, see the Methods section Analysis of mitochondrial mutations and reconstruction of clonal hierarchies. Take note that while the order of mutations is based on the PhISCS model, the grouping of mutations into clones is based on an arbitrary cutoff to provide a useful clustering for further analyses. See Clone in Fig. 2c legend for color codes. f Like e, but for P2.

Back to article page