Wang Ling-Yu, Zhang Lu-Qiang, Li Qian-Zhong, Bai Hui
Laboratory of Theoretical Biophysics, School of Physical Science and Technology, Inner Mongolia University, Hohhot 010021, China.
The State Key Laboratory of Reproductive Regulation and Breeding of Grassland Livestock, Inner Mongolia University, Hohhot 010070, China.
Biophys Rep. 2023 Feb 28;9(1):45-56. doi: 10.52601/bpr.2023.220022.
Abnormal histone modifications (HMs) can promote the occurrence of breast cancer. To elucidate the relationship between HMs and gene expression, we analyzed HM binding patterns and calculated their signal changes between breast tumor cells and normal cells. On this basis, the influences of HM signal changes on the expression changes of breast cancer-related genes were estimated by three different methods. The results showed that H3K79me2 and H3K36me3 may contribute more to gene expression changes. Subsequently, 2109 genes with differential H3K79me2 or H3K36me3 levels during cancerogenesis were identified by the Shannon entropy and submitted to perform functional enrichment analyses. Enrichment analyses displayed that these genes were involved in pathways in cancer, human papillomavirus infection, and viral carcinogenesis. Univariate Cox, LASSO, and multivariate Cox regression analyses were then adopted, and nine potential breast cancer-related driver genes were extracted from the genes with differential H3K79me2/H3K36me3 levels in the TCGA cohort. To facilitate the application, the expression levels of nine driver genes were transformed into a risk score model, and its robustness was tested via time-dependent receiver operating characteristic curves in the TCGA dataset and an independent GEO dataset. At last, the distribution levels of H3K79me2 and H3K36me3 in the nine driver genes were reanalyzed in the two cell lines and the regions with significant signal changes were located.
异常的组蛋白修饰（HMs）可促进乳腺癌的发生。为了阐明HMs与基因表达之间的关系，我们分析了HM结合模式，并计算了乳腺肿瘤细胞与正常细胞之间的信号变化。在此基础上，通过三种不同方法评估了HM信号变化对乳腺癌相关基因表达变化的影响。结果表明，H3K79me2和H3K36me3可能对基因表达变化的贡献更大。随后，通过香农熵鉴定了2109个在癌症发生过程中H3K79me2或H3K36me3水平存在差异的基因，并进行了功能富集分析。富集分析显示，这些基因参与了癌症、人乳头瘤病毒感染和病毒致癌等通路。然后采用单变量Cox、LASSO和多变量Cox回归分析，从TCGA队列中具有差异H3K79me2/H3K36me3水平的基因中提取出9个潜在的乳腺癌相关驱动基因。为便于应用，将9个驱动基因的表达水平转化为风险评分模型，并通过TCGA数据集和一个独立的GEO数据集中的时间依赖受试者工作特征曲线测试其稳健性。最后，在两种细胞系中重新分析了9个驱动基因中H3K79me2和H3K36me3的分布水平，并定位了信号变化显著的区域。
