论文部分内容阅读
目的:进一步了解与宫颈癌相关基因的功能以及作用关系。方法:根据Hela基因周期表达数据的时序特点,提出了KLCC聚类法。该方法以K-means法为平台,以基因间的局部相关系数(Local Correlative Coefficient,LCC)为相关性测度,并对K-means法进行了相应的改进,根据基因间的相关关系,分批对基因聚类。结果:在对Hela基因周期表达数据的聚类分析中,得到9个显著的功能类,其中有3个与肿瘤紧密相关,具有优于当前常用算法的性能。结论:KLCC法能够有效地识别Hela基因周期表达数据中的局部相关和异步相关,并对其进行功能显著的聚类,为宫颈癌的基因治疗提供参考和依据。
Objective: To further understand the function and role of cervical cancer-related genes. Methods: According to the temporal characteristics of Hela gene expression data, a KLCC clustering method was proposed. This method uses the K-means method as the platform and the Local Correlative Coefficient (LCC) as the measure of correlation, and improves the K-means method. According to the correlation between genes, Gene clustering. Results: In the cluster analysis of Hela gene expression data, 9 significant functional classes were obtained, of which 3 were closely related to the tumor and had better performance than the current commonly used algorithms. Conclusion: The KLCC method can effectively identify the local correlation and asynchronous correlation in the Hela expression data, and carry out a significant clustering of the functions, providing a reference and basis for the gene therapy of cervical cancer.