Figure 2 | Scientific Reports

Figure 2

From: Machine learning identifies SLC6A14 as a novel biomarker promoting the proliferation and metastasis of pancreatic cancer via Wnt/β-catenin signaling

Figure 2

The process of screening potential diagnostic markers for the PC diagnosis. (A) The LASSO model’s tuning and feature selection process. (B) A graphic illustrating the biomarkers that were chosen using the SVM-RFE method. (C) Venn diagram demonstrating eight diagnostic markers shared by the above methods, including LAMC2, LAMB3, CTSE, SLC6A14, POSTN, CEACAM5, ITGB6 and IAPP. (D) The expression pattern of LAMC2, LAMB3, CTSE, SLC6A14, POSTN, CEACAM5, ITGB6 and IAPP in PC specimens and non-tumor specimens from GSE62452 and GSE28735 datasets. (E) ROC assays were applied to examine the diagnostic value of the above eight genes in screen PC specimens from non-tumor specimens.

Back to article page