Table_1_Identification and validation of hub genes and molecular classifications associated with chronic myeloid leukemia.docx

BackgroundChronic myeloid leukemia (CML) is a kind of malignant blood tumor, which is prone to drug resistance and relapse. This study aimed to identify novel diagnostic and therapeutic targets for CML.
MethodsDifferentially expressed genes (DEGs) were obtained by differential analysis of the CML cohort in the GEO database. Weighted gene co-expression network analysis (WGCNA) was used to identify CML-related co-expressed genes. Least absolute shrinkage and selection operator (LASSO) regression analysis was used to screen hub genes and construct a risk score model based on hub genes. Consensus clustering algorithm was used for the identification of molecular subtypes. Clinical samples and in vitro experiments were used to verify the expression and biological function of hub genes.
ResultsA total of 378 DEGs were identified by differential analysis. 369 CML-related genes were identified by WGCNA analysis, which were mainly enriched in metabolism-related signaling pathways. In addition, CML-related genes are mainly involved in immune regulation and anti-tumor immunity, suggesting that CML has some immunodeficiency. Immune infiltration analysis confirmed the reduced infiltration of immune killer cells such as CD8+ T cells in CML samples. 6 hub genes (LINC01268, NME8, DMXL2, CXXC5, SCD and FBN1) were identified by LASSO regression analysis. The receiver operating characteristic (ROC) curve confirmed the high diagnostic value of the hub genes in the analysis and validation cohorts, and the risk score model further improved the diagnostic accuracy. hub genes were also associated with cell proliferation, cycle, and metabolic pathway activity. Two molecular subtypes, Cluster A and Cluster B, were identified based on hub gene expression. Cluster B has a lower risk score, higher levels of CD8+ T cell and activated dendritic cell infiltration, and immune checkpoint expression, and is more sensitive to commonly used tyrosine kinase inhibitors. Finally, our clinical samples validated the expression and diagnostic efficacy of hub genes, and the knockdown of LINC01268 inhibited the proliferation of CML cells, and promoted apoptosis.
ConclusionThrough WGCNA analysis and LASSO regression analysis, our study provides a new target for CML diagnosis and treatment, and provides a basis for further CML research.

背景 慢性髓系白血病（chronic myeloid leukemia, CML）是一类易于发生耐药与复发的恶性血液肿瘤，本研究旨在探寻慢性髓系白血病新型诊断及治疗靶点。
方法 本研究通过对基因表达综合数据库（Gene Expression Omnibus, GEO）中的慢性髓系白血病队列进行差异分析，获取差异表达基因（differentially expressed genes, DEGs）；采用加权基因共表达网络分析（Weighted Gene Co-expression Network Analysis, WGCNA）筛选与慢性髓系白血病相关的共表达基因；借助最小绝对收缩和选择算子（Least Absolute Shrinkage and Selection Operator, LASSO）回归分析筛选核心基因（hub genes），并构建基于核心基因的风险评分模型；采用一致性聚类算法识别慢性髓系白血病的分子亚型；最后通过临床样本与体外实验验证核心基因的表达水平及生物学功能。
结果 差异分析共筛选得到378个差异表达基因；通过WGCNA分析鉴定出369个慢性髓系白血病相关基因，这些基因主要富集于代谢相关信号通路。此外，慢性髓系白血病相关基因主要参与免疫调控及抗肿瘤免疫过程，提示慢性髓系白血病存在一定程度的免疫缺陷。免疫浸润分析证实，慢性髓系白血病样本中CD8+ T细胞等免疫杀伤细胞的浸润水平显著降低。经LASSO回归分析筛选得到6个核心基因：LINC01268、NME8、DMXL2、CXXC5、SCD及FBN1。受试者工作特征（Receiver Operating Characteristic, ROC）曲线证实，核心基因在分析队列与验证队列中均具有较高的诊断价值，而风险评分模型进一步提升了诊断准确率。核心基因还与细胞增殖、细胞周期及代谢通路活性密切相关。基于核心基因表达模式，本研究鉴定出两种分子亚型：Cluster A与Cluster B。其中Cluster B的风险评分更低，CD8+ T细胞、活化树突状细胞浸润水平及免疫检查点（immune checkpoint）表达水平更高，对临床常用的酪氨酸激酶抑制剂（tyrosine kinase inhibitor）敏感性更强。最后，本研究通过临床样本验证了核心基因的表达水平与诊断效能，且敲低LINC01268可抑制慢性髓系白血病细胞增殖并促进细胞凋亡（apoptosis）。
结论 本研究通过WGCNA分析与LASSO回归分析，为慢性髓系白血病的诊断与治疗提供了新型靶点，同时为慢性髓系白血病的后续研究奠定了基础。