Zhu Y, Gan X, Qin R, Lin Z. Identification of Six Diagnostic Biomarkers for Chronic Lymphocytic Leukemia Based on Machine Learning Algorithms.
JOURNAL OF ONCOLOGY 2022;
2022:3652107. [PMID:
36467501 PMCID:
PMC9715328 DOI:
10.1155/2022/3652107]
[Citation(s) in RCA: 2] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Subscribe] [Scholar Register] [Received: 07/17/2022] [Revised: 09/13/2022] [Accepted: 09/30/2022] [Indexed: 09/19/2023]
Abstract
BACKGROUND
Chronic lymphocytic leukemia (CLL) is the most common type of leukemia in adults. Thus, novel reliable biomarkers need to be further explored to increase diagnostic, therapeutic, and prognostic effectiveness.
METHODS
Six datasets containing CLL and control samples were downloaded from the Gene Expression Omnibus database. Differential gene expression analysis, weighted gene coexpression network analysis (WGCNA), and the least absolute shrinkage and selection operator (LASSO) regression were applied to identify potential diagnostic biomarkers for CLL using R software. The diagnostic performance of the hub genes was then measured by the receiver operating characteristic (ROC) curve analysis. Functional analysis was implemented to uncover the underlying mechanisms. Additionally, correlation analysis was performed to assess the relationship between the hub genes and immunity characteristics.
RESULTS
A total number of 47 differentially expressed genes (DEGs) and 25 candidate hub genes were extracted through differential gene expression analysis and WGCNA, respectively. Based on the 14 overlapped genes between the DEGs and the candidate hub genes, LASSO regression analysis was used, which identified a final number of six hub genes as potential biomarkers for CLL: ABCA6, CCDC88A, PMEPA1, EBF1, FILIP1L, and TEAD2. The ROC curves of the six genes showed reliable predictive ability in the training and validation cohorts, with all area under the curve (AUC) values over 0.80. Functional analysis revealed an abnormal immune status in the CLL patients. A significant correlation was found between the hub genes and the immune-related pathways, indicating a possible tight connection between the hub genes and tumor immunity in CLL.
CONCLUSION
This study was based on machine learning algorithms, and we identified six genes that could be possible CLL markers, which may be involved in CLL pathogenesis and progression through immune-related signal pathways.
Collapse