Methylation and transcriptome analysis reveal lung adenocarcinoma-specific diagnostic biomarkers

• Published in: Journal of translational medicine Vol. 17; no. 1; pp. 324 - 14
• Main Authors: Li, Rui, Yang, Yi-E, Yin, Yun-Hong, Zhang, Meng-Yu, Li, Hao, Qu, Yi-Qing
• Format: Journal Article
• Language: English
• Published: England BioMed Central Ltd 27.09.2019; BioMed Central; BMC
• Subjects: Adenocarcinoma; Bioinformatics; Biological markers; Biomarkers; Computational biology; Deoxyribonucleic acid; Diagnosis; DNA; DNA methylation; Epigenetics; Gene expression; Genes; Genetic aspects; Genetic markers; Genomes; Health aspects; Lung adenocarcinoma; Lung cancer; Medical prognosis; Messenger RNA; Methylation; Methylation-driven lncRNA; Methylation-driven mRNA; Overall survival rate; Prediction models; Prognosis; Regression analysis; RNA; Software; Survival; Survival analysis; Transcription (Genetics); Yu Zhang; China; Biomarkers; Overall survival rate; Lung adenocarcinoma; Methylation-driven mRNA; Methylation-driven lncRNA
• ISSN: 1479-5876; 1479-5876
• DOI: 10.1186/s12967-019-2068-z

DNA methylation can regulate the role of long noncoding RNAs (lncRNAs) in the development of lung adenocarcinoma (LUAD). The present study aimed to identify methylation-driven lncRNAs and mRNAs as biomarkers in the prognosis of LUAD using bioinformatics analysis. Differentially expressed RNAs were obtained using the edge R package from 535 LUAD tissues and 59 adjacent non-LUAD tissues. Differentially methylated genes were obtained using the limma R package from 475 LUAD tissues and 32 adjacent non-LUAD tissues. Methylation-driven mRNA and lncRNA were obtained using the MethylMix R package from 465 LUAD tissues with matched DNA methylation and RNA expression and 32 non-LUAD tissues with DNA methylation. Gene ontology and ConsensusPathDB pathway analysis were performed to identify functional enrichment of methylation-driven mRNAs. Univariate and multivariate Cox regression analyses were performed to identify the independent effect of each variable for predicting the prognosis of LUAD. Kaplan-Meier curve analysis of DNA methylation and gene expression might provide potential prognostic biomarkers for LUAD patients. A total of 99 methylation-driven mRNAs and 17 methylation-driven lncRNAs were obtained. Univariate and multivariate Cox regression analysis showed that 6 lncRNAs (FOXE1, HOXB13-AS1_2, VMO1, HIST1H3F, AJ003147.8, ASXL3) were retrieved to construct a predictive model associated with overall survival in LUAD patients. Combined DNA methylation and gene expression survival analysis revealed that 4 lncRNAs (AC023824.1, AF186192.1, LINC01354 and WASIR2) and 8 mRNAs (S1PR1, CCDC181, F2RL1, EFS, KLHDC9, MPV17L, GKN2, ITPRIPL1) might act as independent biomarkers for the prognosis of LUAD. Methylation-driven lncRNA and mRNA contribute to the survival of LUAD, and 4 lncRNAs and 8 mRNAs might be potential biomarkers for the prognosis of LUAD.