Fig. 1

The ECM is significantly dysregulated in tumor compared with non-tumor tissue in SqCC. A Workflow describing the approach to characterizing the ECM landscape in SqCC. B Principal component analysis of core matrisomal gene expression in tumor (T) compared with non-tumor (NT) tissue. C Distribution of differentially expressed genes between tumor and non-tumor. 59.5% of matrisome-associated (488 genes) and 62.0% of core matrisomal (170) genes are differentially expressed in SqCC tumors. This represents 59.1% (26 genes), 57.1% (20 genes), and 63.6% (124 genes) of collagens, proteoglycans, and glycoproteins, respectively. It also represents 63.7% (109 genes), 64.3% (153 genes), and 54.1% (186 genes) of all ECM-affiliated proteins, ECM regulators, and secreted factors, respectively. Number of genes indicated in brackets. D Correlation analysis of differentially expressed core matrisomal genes identifies four major clusters of genes in SqCC tumors. r =Spearman correlation coefficient. Non-significant correlations are colored white; positively correlated genes are red and negatively correlated genes are blue. E Expression score of core matrisomal genes corresponding to the four correlational clusters shown in D, in specific lung cell types separated by tumor (T) and non-tumor (N) tissue. EC: endothelial cell