Integration of Bulk and Single-Cell RNA-Seq Data to Construct a Prognostic Model of Membrane Tension-Related Genes for Colon Cancer

Abstract

Background: The plasma membrane provides a highly dynamic barrier for cancer cells to interact with their surrounding microenvironment. Membrane tension, a pivotal physical property of the plasma membrane, has attracted widespread attention since it plays a role in the progression of various cancers. This study aimed to identify a prognostic signature in colon cancer from membrane tension-related genes (MTRGs) and explore its implications for the disease. Methods: Bulk RNA-seq data were obtained from The Cancer Genome Atlas (TCGA) database, and then applied to the differentially expressed gene analysis. By implementing a univariate Cox regression and a LASSO-Cox regression, we developed a prognostic model based on four MTRGs. The prognostic efficacy of this model was evaluated in combination with a Kaplan–Meier analysis and receiver operating characteristic (ROC) curve analysis. Moreover, the relationships between the signature and immune cell infiltration, immune status, and somatic mutation were further explored. Lastly, by utilizing single-cell RNA-seq data, cell type annotation, pseudo-time analysis, drug sensitivity, and molecular docking were implemented. Results: We constructed a 4-MTRG signature. The risk score derived from the model was further validated as an independent variable for survival prediction. Two risk groups were divided based on the risk score calculated by the 4-MTRG signature. In addition, we observed a significant difference in immune cell infiltration, such as subsets of CD4 T cells and macrophages, between the high- and low-risk groups. Moreover, in the pseudo-time analysis, TIMP1 was found to be more highly expressed with the progression of time. Finally, three small molecule drugs, elesclomol, shikonin, and bryostatin-1, exhibited a binding potential to TIMP-1. Conclusions: The novel 4-MTRG signature is a promising biomarker in predicting clinical outcomes for colon cancer patients, and TIMP1, a member of the signature, may be a sensitive regulator of the progression of colon cancer.

Qi Lang formula relieves constipation via targeting SCF/c-kit signaling pathway: An integrated study of network pharmacology and experimental validation

“…After manual merging and annotation, we ultimately presented 10 unique cell clusters ( Fig. 3 A) [ 23 , 24 ]. Then, we annotated the clusters based on several canonical marker genes for known cell lineages: T cell (CD3D, CD3E), ICC (ANO1), smooth muscle (MYH11), lymphatic endothelial cell (CCL21), glial cell (S100B), PDGFRA + cell (PDGFRA), macrophages (CD14), mast cell (TPSAB1, TPSB2), Vessel endothelial cell (VWF) and unknow ( Fig.…”

Integrative analysis reveals a four-gene signature for predicting survival and immunotherapy response in colon cancer patients using bulk and single-cell RNA-seq data

“…scRNA-seq, an innovative methodology, allows us to precisely distinguish different cell types within tissues, providing a powerful tool to understand the complex mix of cell populations when comparing, for instance, tumor and normal tissues (18,20). Additionally, scRNA-seq has the potential to uncover crucial hub genes related to tumor initiation and cancer advancement, which could be instrumental in shaping personalized therapies for CC patients (21)(22)(23)(24)(25)(26)36).…”

“…Recent advances in scRNA-seq help categorize colorectal cancer cells, explore gene differences, and distinguish between primary tumors and metastases (18)(19)(20). A few studies have effectively combined scRNA-seq and bulkRNA-seq data to establish and validate prognostic signatures in CC (21)(22)(23)(24)(25)(26), such as identifying genes related to membrane tension and aging-related or autophagy-related genes (21)(22)(23).…”

Membrane tension-mediated stiff and soft tumor subtypes closely associated with prognosis for prostate cancer patients

“…For the validation set, we used 2 gene expression omnibus (GEO) datasets (GSE46602 [ 35 ], GSE116918 [ 36 ]) and MSKCC2010 [ 37 ] ( http://www.cbioportal.org/ ). Moreover, we obtained total 44 membrane-related genes from previously published literature [ 38 ]. Subsequently, we performed differential analysis between tumor tissue and normal tissue within the TCGA cohort based on R package “limma”.…”