Fig. 2: The SNP rs6465133 in SRI has the potential to distinguish PCAND from T2DM via machine learning. | Experimental & Molecular Medicine

Fig. 2: The SNP rs6465133 in SRI has the potential to distinguish PCAND from T2DM via machine learning.

From: Sorcin can trigger pancreatic cancer-associated new-onset diabetes through the secretion of inflammatory cytokines such as serpin E1 and CCL5

Fig. 2

a SHAP values of the 15 SNPs with the greatest significance in distinguishing PCAND from T2DM. b The corresponding genes of the top 15 SNPs. The expression patterns of the (c) SRI and (d) STK11 genes in PC tumors (T) compared with those in normal pancreas tissue (N). e The expression pattern of SRI in PC patients with different pathological TNM stages. Comparison of (f) overall survival rates and (g) disease-free survival rates between the high-SRI group and the low-SRI group, with the median SRI expression as the cutoff. Fig. c-g are based on The Cancer Genome Atlas (TCGA) database and Genotype-Tissue Expression (GTEx). h Statistical analysis of the sorcin level in pancreatic cancer tissues and adjacent normal tissues from 88 patients diagnosed with PDAC. i Statistical analysis of sorcin levels in pancreatic cancer tissues from 88 PDAC patients diagnosed with different pathological TNM stages. j Comparison of overall survival rates between the high SRI group (with scores ranging from 712) and the low SRI group (with scores ranging from 06). h–j are based on PDAC patients enrolled at the Second Affiliated Hospital of Zhejiang University between January 2013 and December 2017. *P < 0.05; ****P < 0.0001, means ± SD are shown. Statistical analysis was performed via Student’s t test for two groups and one-way ANOVA for multiple groups. Kaplan–Meier curves of survival were compared via the log-rank test.

Back to article page