ObjectiveTo study the abnormal biological pathways of intrahepatic cholangiocarcinoma (ICC) from the transcriptomics level and identify genes associated with the prognosis of ICC.MethodsThe differentially expressed genes were screened by t test and fold change method, then KEGG functional enrichment analysis was performed on related genes. The STRING database was applied to construct protein interaction network and find the hub nodes of the network by calculating the degree, betweenness, and closeness of each node. Kaplan-Meier survival analysis was performed using log-rank test to identify prognostic genes related to ICC.ResultsAll of 1 134 differentially expressed genes were overlapped in 3 datasets, which were mainly involved in 15 pathways, including DNA replication, cell cycle, drug metabolism, RNA transport, etc. signaling pathways and amino acid synthesis. According to protein interaction network analysis, TAF1, GRB2, E2F4, HNF4A, MYC, and TP53 genes were hub nodes. As GRB2 and TP53 genes were also the death related genes of ICC, it was found that patients with lower GRB2 gene expression had a better overall survival than those with higher GRB2 gene expression (P=0.040 9), while patients with lower TP53 had a worse overall survival than those with higher TP53 gene expression (P=0.027 3), which were also verified in the TCGA database.ConclusionsThe abnormal cell metabolism is notably related to the tumorigenesis of ICC. TAF1, GRB2, E2F4, HNF4A, MYC, and TP53 are the key genes in the carcinogenesis and progression of ICC. Expressions of GRB2 and TP53 genes are associated with the prognosis of ICC.
Due to the high dimensionality and complexity of the data, the analysis of spatial transcriptome data has been a challenging problem. Meanwhile, cluster analysis is the core issue of the analysis of spatial transcriptome data. In this article, a deep learning approach is proposed based on graph attention networks for clustering analysis of spatial transcriptome data. Our method first enhances the spatial transcriptome data, then uses graph attention networks to extract features from nodes, and finally uses the Leiden algorithm for clustering analysis. Compared with the traditional non-spatial and spatial clustering methods, our method has better performance in data analysis through the clustering evaluation index. The experimental results show that the proposed method can effectively cluster spatial transcriptome data and identify different spatial domains, which provides a new tool for studying spatial transcriptome data.