Categories
Nevin Manimala Statistics

Construction of Prognostic Model and miRNA-mRNA Regulatory Network for Lung Squamous Cell Carcinoma

Altern Ther Health Med. 2023 Oct 20:AT9323. Online ahead of print.

ABSTRACT

The purpose of this paper was to construct a prognostic model, miRNA-mRNA regulatory network and protein-protein interaction (PPI) network for lung squamous cell carcinoma (LUSC) used data from the cancer genome atlas (TCGA) database. In this study, we first downloaded and sorted out the expression matrix containing 19962 mRNA transcripts (including 502 LUSC and 51 normal control (NC) samples) and the expression matrix containing 2205 miRNA transcripts (including 478 LUSC and 45 NC samples) from the TCGA database. We obtained 389 differentially expressed miRNAs (DE-miRNAs), of which 305 were upregulated and 84 down-regulated DE-miRNAs. Next, a total of 7 prognosis-related DE-miRNAs (PDE-miRNAs) were identified by Cox regression analysis, and the prognosis model consisting of three PDE-miRNAs (hsa-miR-4746-5p, hsa-miR-556-3p and hsa-miR-489-3p) was optimized. Then, we drew the survival curves and found that the survival rates of the three PDE-miRNA high and low expression groups and the survival rates of the high-risk and low-risk patients in the prognosis model had significant statistical differences. In addition, the receiver operating characteristics (ROC) curve analysis and independent prognostic analysis confirmed that the prognostic model we built has a relatively accurate ability to predict the grouping and prognosis of LUSC patients. Finally, Cox regression analysis were used to construct the miRNA-mRNA regulatory network, which showed the regulatory relationship between PDE-miRNAs and targeted mRNAs. Moreover, we constructed the PPI network composed of 145 targeted mRNAs and the subnetwork composed of 10 hub-targeted mRNAs (FCGR3A, IL13, CCR2, PPARGC1A, FCGR3B, ACSL1, PLXNA4, LPL, KAT2B and AOC3), which showed the interaction between targeted mRNAs. The above results indicated that the prognosis model we built can predict LUSC patients relatively accurately. The miRNA-mRNA regulatory network and the PPI network of targeted mRNAs illustrated the regulatory mechanisms and interactions between RNAs, which were of certain reference significance for us to further understand the molecular pathogenesis of LUSC and for clinical early diagnosis and treatment.

PMID:37856813

By Nevin Manimala

Portfolio Website for Nevin Manimala