Exploratory screening of potential pan-cancer biomarkers based on The Cancer Genome Atlas database.
- Author:
Chuan ZHOU
1
;
Xue MA
1
;
Yun Kun XING
1
;
Lu Di LI
1
;
Jie CHEN
1
;
Bi Yun YAO
1
;
Juan Ling FU
1
;
Peng ZHAO
1
Author Information
1. Department of Toxicology, Beijing Key Laboratory of Toxicological Research and Risk Assessment for Food Safety, Peking University School of Public Health, Beijing 100191, China.
- Publication Type:Journal Article
- Keywords:
Biomarkers, tumor;
Gene expression regulation;
Genome, human;
Pan-cancer
- MeSH:
Biomarkers, Tumor/genetics*;
Early Detection of Cancer;
Gene Expression Profiling;
Gene Expression Regulation, Neoplastic;
Humans;
MicroRNAs/genetics*;
Neoplasms/genetics*;
Prognosis
- From:
Journal of Peking University(Health Sciences)
2021;53(3):602-607
- CountryChina
- Language:Chinese
-
Abstract:
OBJECTIVE:To screen potential pan-cancer biomarkers based on The Cancer Genome Atlas (TCGA) database, and to provide help for the diagnosis and prognosis assessment of a variety of cancers.
METHODS:"GDC Data Transfer Tool" and "GDCRNATools" packages were used to obtain TCGA database. After data sorting, a total of 13 cancers were selected for further analysis. False disco-very rate (FDR) < 0.05 and fold change (FC) >1.5 were used as the differential expression criteria to screen genes and miRNAs that were up- or down-regulated in all the 13 cancers. In the receiver operating characteristic curve (ROC curve), the area under the curve (AUC), the best cut-off value and the corresponding sensitivity and specificity were used to reflect diagnostic significance. The Kaplan-Meier method was used to calculate the survival probability and then the log-rank test was performed. Hazard ratio (HR) was calculated to reflect prognostic evaluation significance. DAVID tool were used to perform GO (Gene Ontology) and KEGG (Kyoto Encyclopedia of Genes and Genomes) enrichment analysis for differentially expressed genes. STRING and TargetScan tools were used to analyze the regulatory network of differentially expressed genes and miRNAs.
RESULTS:A total of 48 genes and 2 miRNAs were differentially expressed in all the 13 cancers. Among them, 25 genes were up-regulated, 23 genes and 2 miRNAs were down-regulated. Most differentially expressed genes and miRNAs had good ability to distinguish between the cases and controls, with AUC, sensitivity and specificity up to 0.8-0.9. Survival analysis results show that differentially expressed genes and miRNAs were significantly associated with patient survival in a variety of cancers. Most up-regulated genes were risk factors for patient survival (HR>1), while most down-regulated genes were protective factors for patient survival (0 < HR < 1). The enrichment analysis of GO and KEGG showed that the differentially expressed genes were mostly enriched in biological events related to cell proliferation. In the regulatory network analysis, a total of 13 differentially expressed genes and 2 differentially expressed miRNAs had regulatory and interaction relationships.
CONCLUSION:The 48 genes and 2 miRNAs that were differentially expressed in 13 cancers may serve as potential pan-cancer biomarkers, providing help for the diagnosis and prognosis evaluation of a variety of cancers, and providing clues for the development of broad-spectrum tumor therapeutic targets.
