LAUSR

Energy metabolism therapy has gradually shown its potential in the treatment of tumor patients, but it has significant selectivity, thus distinguishing energy subtypes of lung adenocarcinoma (LUAD) is necessary to identify patients who may benefit from energy metabolism interference therapy. Gene expression data downloaded from The Cancer Genome Atlas and Gene Expression Omnibus, molecular subtypes were selected using NMF algorithm, prognostic differentially expressed genes (DEGs) were identified with DESeq. 2 and survival package, Lasso and cox regression analysis were used to Construct of Risk Signature. The relationship between molecular subtypes and prognosis as well as clinical characteristics were evaluated. Univariate and multivariate COX regression were used to analyze the correlation between the signature and patient prognosis. Based on 592 energy metabolism‐related genes, 430 LUAD samples were divided into three subtypes, of which C2 has the worst prognosis, and 942 prognostic DEGs were identified. 11‐gene prognostic risk signature was constructed. Compared with the traditional clinical features of T, N, and age, this 11‐gene signature performs better in predicting the risk of LUAD prognosis. At the same time, it is an independent risk factor for patient prognosis. The signature showed strong robustness in different cohorts. Compared with other published signatures, 11‐gene signatures have strong clinical applicability and accuracy. The predictive signature will enable patients with LUAD to be more accurately managed in clinical practice.