Background: Gastric cancer (GC) is the fifth most frequently diagnosed malignancy, and the third leading cause of tumor-related mortalities worldwide. Due to a high heterogeneity in GC, its treatment and prognosis are challenging, necessitating urgent identification of novel prognostic predictors for GC patients. Methods: We downloaded RNA sequence data, from the Cancer Genome Atlas and microarray data from Gene Expression Omnibus database, then identified common differentially-expressed genes (DEGs) between GC and normal gastric tissues across four datasets. We then used a combination of protein-protein interaction (PPI) network and weighted gene co-expression network analysis (WGCNA) to identify key genes with prognostic value in GC. Thereafter, we used quantitative real time polymerase chain reaction (qRT-PCR) to validate expression of the identified key genes in the Zhejiang University (ZJU) cohort. Finally, we evaluated the relationships between gene expression and immune factors, including immune cells and biomarkers of immunotherapy. Results: Among 426 common DEGs screened, 333 and 93 were upregulated and downregulated, respectively. PPI network and WGCNA successfully identified the top 30 hub genes, among which PTPRC, TYROBP, CCR1, CYBB, LCP2, and C1QB were common. Furthermore, TYROBP and C1QB were negatively associated with prognosis of GC patients, implying that they were key GC predictors. Interestingly, TYROBP and C1QB were positively correlated with predictive biomarkers for GC immunotherapy, including PD-L1 expression, CD8 + T cells infiltration, and EBV status. Conclusions: TYROBP and C1QB were identified as two novel key genes with prognostic value in GC by network analysis.