Identification of different subtypes of ovarian cancer and construction of prognostic models based on glutamine-metabolism associated genes

Abstract

Ovarian cancer (OC) is common malignant tumor of female reproductive system. Glutamine metabolism-related genes (GMRGs) play a key role in ovarian cancer. Here, available database-- The Cancer Genome Atlas (TCGA), Genotype-Tissue Expression (GTEx), and Gene Expression Omnibus (GEO) databases were applied in our research. OC samples from TCGA were divided into different clusters based on Cox analysis, which filtering GMRGs with survival information. Then, differentially expressed genes (DEGs) between these clusters were intersected with DEGs between normal ovary samples and OC samples, and GMRGs in order to obtain GMRGs-related DEGs. Next, a risk model of OC was constructed and enrichment analysis of risk model was performed based on hallmark gene set. Besides, the immune cells ratio in OC samples were detected via Cell type Identification By Estimating Relative Subsets Of RNA Transcripts (CIBERSORT). Finally, we explored a series of potential biomarkers of OC. In this research, 9 GMRGs-related DEGs were obtained. GMRGs-related DEGs were enriched to canonical Wnt signaling pathway.NKD2, C2orf88, and KLHDC8A, which were significantly associated with prognosis, were retained for risk model construction. Based on the risk model, 18 hallmark pathways with significant difference were enriched. Fifteen types of immune cells (such as iDC, NK CD56dim cells, and neutrophils) enjoying significant difference between these 2 risk groups (high risk group vs. low risk group) were detected, which indicates possible disparate TME in different metabolic subtypes of ovarian cancer.