Extended Data Fig. 4: Characteristics of myeloid cells.
Related to Fig. 2. (a) Line plots depicting the average silhouette width of myeloid cells across a range of resolutions from 0.1 to 1.5. (b) Clustering trees of the myeloid cells colored according to the expression of known markers. The node colors indicate the average of the log2 TPM of samples in each cluster. CLEC9A shows a population of conventional type 1 dendritic cells (cDC1), and C1QC is a marker of macrophage cells. (c) Box plot illustrating cell purity for each myeloid cell type, calculated using ROGUE from 797 samples. The bottom of each box indicates the Q1, and the top represents the Q3. The height of the box reflects the IQR, and the horizontal line inside the box indicates the median. The whiskers extend to the positions of Q1 - 1.5 * IQR and Q3 + 1.5 * IQR. (d) Dot plot depicting the expression of representative marker genes of each myeloid cell type. (e) Box plot showing the proportions of distinct dendritic cell subgroups across different sources, including blood (red), normal tissue (green), precancerous (orange), tumor (purple), and metastatic (blue), derived from 496 treatment-naïve samples. Significance labels in the figure was assessed via Kruskal-Wallis tests to compare each subgroup distributions among five tissue types. Significance for pairwise source proportions within each cell type, assessed via two-tailed unpaired Wilcoxon tests, is reported in Supplementary Table 6. The open rectangle annotates the comparative scope, with BH correction for multiple testing. The bottom of the box represents the Q1, and the top of the box represents the Q3. The height of the box represents the IQR, while the horizontal line inside the box indicates the median. The whiskers extend to the positions of Q1 - 1.5 * IQR and Q3 + 1.5 * IQR. (f) Heatmap showing the ORs of myeloid cell sub types occurring in each source. OR > 1.5 indicates that the cell type is preferred to distribute in the corresponding source. Significance was assessed using the two-sided Fisher test and adjusted using the BH method. (g) Heatmap showing the enriched pathways for each myeloid cell subset. (h) Heatmap showing the number of overleaped differentially expressed genes among all cell types. (i) Heatmap depicting the correlation of cell type signature scores calculated by GSVA between before and after corrected in TCGA-ESCA dataset.
