CAJ | 학술논문

通过对基因表达谱数据的分析从而促进肿瘤诊断与治疗技术的发展，其研究正成为生物医学领域的一个热点。因此，提出了一种熵信息处理和主成分分析（ principal component analysis ， PCA ）相结合的方法。首先运用熵信息对超高维基因表达谱数据进行粗选取，得到特征基因子集；由于基因子集仍存在相关性，进而利用PCA对其进一步冗余剔除；最后对得到的无冗余且具有正交性信息的基因特征进行真实数据实验。实验结果显示所采用的方法能有效去除肿瘤样本中的不相关和冗余信息，同时最大程度的保留肿瘤分类信息。与其他肿瘤分类方法相比，在精度上具有比较明显的优势，从而验证了该方法是有效的、可行的。
통과대기인표체보수거적분석종이촉진종류진단여치료기술적발전，기연구정성위생물의학영역적일개열점。인차，제출료일충적신식처리화주성분분석（ principal component analysis ， PCA ）상결합적방법。수선운용적신식대초고유기인표체보수거진행조선취，득도특정기인자집；유우기인자집잉존재상관성，진이이용PCA대기진일보용여척제；최후대득도적무용여차구유정교성신식적기인특정진행진실수거실험。실험결과현시소채용적방법능유효거제종류양본중적불상관화용여신식，동시최대정도적보류종류분류신식。여기타종류분류방법상비，재정도상구유비교명현적우세，종이험증료해방법시유효적、가행적。
Analysis of the gene expression data to promote the technology of cancer diagnostic and treatment is becoming a hot research in today′s field of bioinformatics .Therefore, this paper presents a method combined entropy of information processing and PCA .The use of entropy information on ultra-high-dimensional gene expression data can coarse selection and get feature subset .Due to the subset of genes are still relevant , the use of PCA can further eliminate its redundancy .The non-redundant information and orthogonal genetic characteristics obtained were real data experiment .The results show that the method used in this paper can effectively rule out redun-dancy of samples while maximum preserve the overall gene information .Compared with other tumor classification method , it has obvi-ous advantages in accuracy , which proves that the method is effective and feasible .