Figure 1. Genetic heterogeneity of immune cells and pseudo-diploid cells in breast cancer biopsies.

(A) Genome-wide copy number view of representative single-cell T-cell genomes illustrating T-Cell Receptor alpha (TCRa) and beta (TCRb) deletions. (B) Genome-wide copy number view of representative single-cell B-cell genomes illustrating light (IGL) and heavy (IGH) immunoglobulin deletions. (C) Bar plot quantification of T-cells, B-cells, pseudo-diploid cells and other non-tumor cells identified in profiled tumor biopsies. (D) Zoomed in views of TCRa deletion breakpoints in single T-cells found in tumors biopsies (P15 and P9), blood purified CD8+/4+ T-cells, and single-leukemic cells derived from a T-cell leukemia. (E) Representative genome-wide copy number plot of a T-cell exhibiting X-chromosome loss. Insert – bar plot quantification of X-loss T-cells in ER+ and ER- tumors. Asterisk denotes statistical significance based on chi-square test (p-value=0.0047). (F) Left panel: Frequency plot of CNAs identified in a panel of 200 breast cancer genomes. Highly recurrent alterations, such as 1q gain and 16 p loss are noted. Right Panel: Representative copy number plots of pseudo-diploid single-cell genomes illustrating the occurrence of recurrent breast cancer CNAs in these cells.

Figure 1—figure supplement 1. Characteristics of tumor samples profiled in the study.

(A) Schematic illustration of single-nuclei sorting strategy employed in the study. (B) Hierarchal clustering heatmap of breast cancer samples illustrating PAM50 subtype information based on differential gene expression analysis. Tumors analyzed in the present study (n = 16) are annotated. (C) Flow cytometric profiles of analyzed tumors illustrating ploidy distributions as measured by DNA content using DAPI staining. Asterisk denotes distribution/ploidy peak where cancer cells were detected in the single-nuclei sequencing data. (D) Illustration of summary statistics: approximate ploidy, tumor cellularity (% cancer), tumor size, and patient age of analyzed tumor samples. Ploidy and cellularity statistics are based on flow cytometry and single-nuclei sequencing data.

Figure 1—figure supplement 2. Single-cell sequencing analysis of T-cells retrieved from purified lymphocytes and T-cells identified in breast cancer tissue biopsies.

(A) Representative genome-wide copy number profiles of normal cells showing non-rearranged ‘flat’ genomes. (B) Zoom-in-view of deletion genomic coordinates at the TCRalpha locus of a representative single T-cell genome. (C) Left panel, bar plot quantification of TCRalpha (TCRa) and TCRbeta (TCRb) deletion frequencies observed in 95 sequenced single CD4+/CD8+ T-cells. Right panel, bar plot illustration of light chain (IGL) and heavy chain (IGH) deletion detection frequencies derived from single-cell sequencing of 93 CD27+/CD13+ B-cells. (D) Breakpoint analysis of TCRa deletions in CD4+/CD8+ T-cells illustrating the heterogeneity of breakpoint occurrence in genomic bins. (E) Zoom-in-view of the heterogeneity of TCRa deletion breakpoints found in X-loss T-cells identified from two representative patients tumors. (F) Pair-wise distances of genomic bins within TCRa deletion breakpoints in single T-cells found in tumors biopsies (P15 and P59), blood purified CD8+/4+ T-cells, and single-leukemic cells derived from T-cell leukemia. (G) Representative genome-wide copy number profiles of T-cells with karyotypic abnormalities. Examples shown are: gain on chromosome 16 and an interstitial deletion on chromosome nine in two different T-cells genomes.

Figure 1—figure supplement 3. Copy number alterations in pseudo-diploid cells identified in tumor samples.

(A) Representative genome-wide copy number profiles of patient matched pseudo-diploid and cancer cells from two patients. Shaded in gray are chromosomes where pseudo-diploid CNAs are identified. (B) Heatmap illustration of copy number profiles of all identified (n = 28) pseudo-diploid single cells in analyzed tumor samples.