Fig. 1

Characterization of LY6G6D and FUT4/CD15 expression. a The work flow on the left shows cancer cell line transcriptomic samples that were retrieved from NCBI (Barretina J et al. 2102) and interrogated for differentially expressed genes of known immune-related genes from ImmPort collection. Right, unsupervised hierarchical cluster of cancer cell lines (n = 604) shows a gene signature enriched in colorectal cancer. Enlarged image shows two genes LY6G6D and FUT4/CD15 within the cluster that are upregulated in Microsatellite stable (MSS) but not in microsatellite instable (MSI) colon cancer cells categorized for mutational load and copy number variations (CNVs). b Quantification of CD15 and LY6G6D mRNA in patient-matched tumor-normal mucosa extracted from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) data sets. Scatter plot in which each circle represents mRNA levels in each tumor sample, horizontal line is the mean value. *P ≤ 0.05; **P ≤ 0.01; ***P ≤ 0.001 by Mann–Whitney U test. c Heat map of log-transformed odds ratios of a set of immune-related genes for two different molecular phenotypes MSI vs MSS. On the left, quantification of LY6G6D mRNA by a box plot in CRCs classified as CIN high or low based on a weighted genome integrity index (see Methods). *P ≤ 0.05; t test Welch-corrected. d Enrichment map network of statistically significant gene interactions. Nodes represent gene hub and lines their connectivity. Node size is proportional to number of line with arrows. Heat map of differentially expressed genes within JAK/STAT and MAPK signaling according to MSI-H, MSI-L, MSS subtypes. Shown are groups with high relative expression (hi, red) versus the low relative expression (lo, blue) at the optimum value cutoff