| 离散增量结合二次判别分析预测人类DNA甲基化位点 |
| |
| 引用本文： | 刘佳,赵秀娟,裴智勇,秦笙,蔡禄.离散增量结合二次判别分析预测人类DNA甲基化位点[J].生物数学学报,2012(3):563-570. |
| |
| 作者姓名： | 刘佳 赵秀娟 裴智勇 秦笙 蔡禄 |
| |
| 作者单位： | [1]内蒙古科技大学数理与生物工程学院,包头内蒙古014010 [2]包头医学院基础学院,包头内蒙古014010 [3]中国科学院植物研究所系统与进化国家重点实验室,北京100093 [4]中国科学院研究生院,北京100039 [5]南京师范大学生命科学学院比较基因组与生物信息学实验室＆江苏省生物多样性与生物技术重点实验室,江苏南京210046 |
| |
| 基金项目： | 国家自然科学基金资助项目（60761001）; 内蒙古自然科学基金（2011MS0504）; 高等学校科学研究基金（NJZY12103） |
| |
| 摘 要： | DNA甲基化作为直接作用于DNA序列的一种表观遗传修饰,能够在不改变DNA分子一级结构的情况下影响基因表达,在生命活动中扮演着重要的角色.在哺乳动物中,DNA甲基化主要发生在C_pG二核苷酸的胞嘧啶上,并且在基因组中呈现不均匀分布.准确预测DNA甲基化位点有助于阐明DNA甲基化对基因表达的调控作用,并为肿瘤的早期诊断及治疗提供新的依据.本文应用离散增量结合二次判别分析的方法,对人类的C_pG二核苷酸甲基化状态进行了识别.5折交叉检验的整体准确率超过了80%,受试者操作特性曲线面积也达到了0.86.与现有方法相比,预测成功率显著提高.这说明离散增量结合二次判别分析方法适用于甲基化位点的预测;基因组序列中甲基化位点具有序列依赖性.
|
| 关 键 词： | DNA甲基化 CpG位点 离散增量 二次判别分析 |
Prediction of human DNA Methylation Sites based on Increment of Diversity Combined with Quadratic Discriminant Analysis |
| |
| LIU Jia,ZHAO Xiu-juan,PEI Zhi-yong,QIN Sheng,CAI Lu.Prediction of human DNA Methylation Sites based on Increment of Diversity Combined with Quadratic Discriminant Analysis[J].Journal of Biomathematics,2012(3):563-570. |
| |
| Authors: | LIU Jia ZHAO Xiu-juan PEI Zhi-yong QIN Sheng CAI Lu |
| |
| Institution: | 1 (1 School of Mathematics,Physics and Biological Engineering,Inner Mongolia University of Science ＆ Technology,Baotou Inner Mongolia 014010 China) (2 School of Foundation,Bao Tou Medical College,Baotou Inner Mongolia 014010 China) (3 State Key Laboratory of Systematic and Evolutionary Botany,Institute of Botany,Chinese Academy of Sciences,Beijing 100093 China) (4 Graduate University of the Chinese Academy of Sciences,Beijing 100039 China) (5 Laboratory for Comparative Genomics and Bioinformatics ＆ Jiangsu Key Laboratory for Biodiversity and Biotechnology,College of Life Science,Nanjing Normal University,Nanjing Jiangsu 210046 China) |
| |
| Abstract: | DNA methylation is a kind of epigenetic modification way of DNA sequences and plays crucial roles in gene regulation.In mammalian,the most common type of DNA modification consists of the methylation of cytosine in the CpG dinucleotide,and the methylated regions within the genome are uneven distribution.Successful prediction of DNA methylation sites can help us explore its roles in gene regulation and provide possible clues for tumor diagnoses and therapy. Accurate identification of DNA methylation sites is substantially important for understanding the pattern of gene expression.In this study,the method of Increment of Diversity combined with Quadratic Discriminant analysis（IDQD） was used to predict the DNA methylation sites. The methylation patterns form various sources ranging from plants to humans.The results of 5-fold cross-validation test show an accuracy of 80%and the area under ROC（auROC） curves of 0.86,which is higher than that of the provious methods.Our data indicate that IDQD can be used in prediction of DNA methylation sites and the patterns of DNA methylation are sequence -dependent. |
| |
| Keywords: | DNA methylation CpG sites Increment of diversity Quadratic discriminant analysis IDQD |
| 本文献已被 CNKI 维普 等数据库收录！ |
|
