Skip to main content
Fig. 4 | Journal of Translational Medicine

Fig. 4

From: Novel method for highly multiplexed gene expression profiling of circulating tumor cells (CTCs) captured from the blood of women with metastatic breast cancer

Fig. 4

Gene expression analysis of Parsortix harvests from HV blood without or with cancer cell spikes A Heatmap of hierarchical clustering for 15 genes in Parsortix harvests from HV blood spiked with either a single cell line cell or 50 pg of cell line RNA. The 15 genes shown are the same as those shown in Fig. 3C. The data were mean centered and scaled, the colors indicate relative Z-scores. The distance method for hierarchical clustering used was non-parametric rank order Kendall correlation and the linkage method used was Ward.D2. The “PTPRC” annotation at the top of the heatmap and the scale of 1 to 8 to the right of the heatmap indicate the level of PTPRC expression in the samples (a range of net Ziplex probe intensities of 565–8200). The correlation coefficient (R) between the expression of individual genes and the WBC marker PTPRC is indicated in the annotations to the right of each gene name. All the nine genes in the bottom gene cluster were well correlated with PTPRC. B Heatmap of hierarchical clustering for 14 potentially informative genes in Parsortix harvests from HV blood spiked with either a single cell line cell or 50 pg of cell line RNA. The 14 genes had at least two-fold-differences between unspiked and at least one spiked sample and had relatively small p-values. The data processing and annotation is the same as in A. The raw data used for hierarchical clustering are in Additional file 8

Back to article page