Research Paper Volume 15, Issue 14 pp 7023—7037

Single-cell transcriptomics reveals the drivers and therapeutic targets of lymph node metastasis in lung adenocarcinoma

class="figure-viewer-img"

Figure 2. Copy number variation analysis of epithelial cells from primary tumor and metastasis lymph node. (A) Representative hierarchical heatmap showing large-scale CNVs in maECs from various samples. Gains (red) or losses (blue) were inferred by averaging the expression over 100 gene stretches on the respective chromosomes. (B) The CNV score facilitated quantitatively determining malignant cells. (C) The summary plot of the CNV profile of malignant ECs from esPT, asPT and asLN. CNVs were converted to the chromosome arm level change and simplified as gain or loss. (DF) The clonal evolutionary trees of maECs from esPT, asPT and asLN, respectively. The length of branch is determined by the percentage of cells containing the corresponding CNV in subclones. (G) The heatmap of GSVA demonstrated differences in enriched pathways among maECs from esPT, asPT and asLN. (H) The hierarchical clustering heatmap of enriched transcription regulon using SCENIC analysis. *, means presenting in >90% of tumor cells. Abbreviations: CNVs: copy number variations; maECs: malignant epithelial cells; esPT: primary tumor in early stage LUAD; esLN: lymph node in early stage LUAD; asPT: primary tumor in advanced stage LUAD; asLN: lymph node in advanced stage LUAD; GSVA: gene set variation analysis; SCEINC: single-cell regulatory network inference and clustering.