Prediction of pancreatic cancer risk pathways based on multiple omics

Abstract

As a common and dangerous malignant tumor, the pathogenesis of pancreatic cancer is still not completely clear. This study aims to reveal the potential molecular mechanisms and therapeutic targets of pancreatic cancer by comprehensively utilizing the protein and gene expression data in TCGA database, combining survival information and protein interaction network analysis, and provide new theoretical support for clinical diagnosis and treatment.

First, we obtained protein and gene expression data from pancreatic cancer patients from the TCGA database, covering protein expression for 328 samples and gene expression information for 493 samples. By incorporating survival information, we obtained a complete expression profile, which lays the foundation for subsequent analysis. Second, 23 differential proteins were identified associated with pancreatic cancer survival that may play an important role in the development and progression of pancreatic cancer. Further, we constructed a protein interaction network for pancreatic cancer, and combined the String database and GN algorithm to identify the proteins with the most significant impact on pancreatic cancer and their associated primary and secondary proteins, providing important clues for further research. Subsequently, we analyzed the gene expression data using the Lasso regression model, identified 6 genes with significant differences (TMEM176A, ANLN, IGFBP2, MROH 9, OLFM 3, TRIM67), and drew their Lasso coefficient pathway maps and cross-validation curves. The random walk algorithm and the perturbation method were further used to identify the two most important pathway genes, SYNE 1 and STARD8, which provided a new perspective on the pathogenesis of pancreatic cancer. Finally, we input SYNE 1 and STARD8 into the KEGG database, and constructed a pathway network of pancreatic cancer, revealing the potential mechanism of action of these two genes in the occurrence and development of pancreatic cancer. These results provide an important theoretical basis for the early diagnosis of pancreatic cancer, the discovery of therapeutic targets, and the development of individualized treatment strategies, and are expected to bring about a significant improvement in the survival rate and quality of life of pancreatic cancer patients.