Weighted gene co-expression network analysis identifies key hub genes and pathways in acute myeloid leukemia

Xinfeng Wang; Akhilesh K Bajpai; Qingqing Gu; David G Ashbrook; Athena Starlard-Davenport; Lu Lu

doi:10.3389/fgene.2023.1009462

Weighted gene co-expression network analysis identifies key hub genes and pathways in acute myeloid leukemia

Front Genet. 2023 Feb 27:14:1009462. doi: 10.3389/fgene.2023.1009462. eCollection 2023.

Authors

Xinfeng Wang¹, Akhilesh K Bajpai², Qingqing Gu^{1

2}, David G Ashbrook², Athena Starlard-Davenport², Lu Lu²

Affiliations

¹ Department of Hematology, Affiliated Hospital of Nantong University, Nantong, Jiangsu, China.
² Department of Genetics, Genomics, and Informatics, University of Tennessee Health Science Center, Memphis, TN, United States.

Abstract

Introduction: Acute myeloid leukemia (AML) is the most common type of leukemia in adults. However, there is a gap in understanding the molecular basis of the disease, partly because key genes associated with AML have not been extensively explored. In the current study, we aimed to identify genes that have strong association with AML based on a cross-species integrative approach. Methods: We used Weighted Gene Co-Expression Network Analysis (WGCNA) to identify co-expressed gene modules significantly correlated with human AML, and further selected the genes exhibiting a significant difference in expression between AML and healthy mouse. Protein-protein interactions, transcription factors, gene function, genetic regulation, and coding sequence variants were integrated to identify key hub genes in AML. Results: The cross-species approach identified a total of 412 genes associated with both human and mouse AML. Enrichment analysis confirmed an association of these genes with hematopoietic and immune-related functions, phenotypes, processes, and pathways. Further, the integrated analysis approach identified a set of important module genes including Nfe2, Trim27, Mef2c, Ets1, Tal1, Foxo1, and Gata1 in AML. Six of these genes (except ETS1) showed significant differential expression between human AML and healthy samples in an independent microarray dataset. All of these genes are known to be involved in immune/hematopoietic functions, and in transcriptional regulation. In addition, Nfe2, Trim27, Mef2c, and Ets1 harbor coding sequence variants, whereas Nfe2 and Trim27 are cis-regulated, making them attractive candidates for validation. Furthermore, subtype-specific analysis of the hub genes in human AML indicated high expression of NFE2 across all the subtypes (M0 through M7) and enriched expression of ETS1, LEF1, GATA1, and TAL1 in M6 and M7 subtypes. A significant correlation between methylation status and expression level was observed for most of these genes in AML patients. Conclusion: Findings from the current study highlight the importance of our cross-species approach in the identification of multiple key candidate genes in AML, which can be further studied to explore their detailed role in leukemia/AML.

Keywords: AML; WGCNA; co-expression; gene expression; leukemia; protein interaction; systems genetics.