摘要
作为目前遗传学热点领域表观遗传学的重要研究课题,核小体定位参与多种生物学过程并起着非常关键的作用,因此用理论方法预测核小体定位具有重要的生物学意义。该工作以酵母基因组为研究对象,根据其核心DNA与连接DNA序列的组分特征差异,分别统计其核苷酸k联体出现的频数,计算其多样性增量,并以此为特征值输入支持向量机构建模型。对酵母核小体核心DNA和连接DNA序列进行分类预测,整体准确率和相关系数分别达到93.10%和0.862。此方法在人类和果蝇的核小体核心DNA和连接DNA序列分类预测中也取得了较为理想的效果。将模型用于预测核小体在基因组上的位置取得初步成功。
Nucleosome positioning is an important issue in epigenetics.The prediction of nucleosome positioning is of great significance as it plays key roles in various biological processes.In this study,a prediction model is constructed using support vector machine,of which input was the increment of diversity based on the difference in k-mer frequency between core DNA and linker DNA.This model was applied to classification prediction of core DNA and linker DNA in S.cerevisiae,the total prediction accuracy and Matthews＇s correlation coefficient are 93.10% and 0.862 respectively.The model also showed good performance in prediction of the nucleosome positioning in H.sapiens and D.melanogaster.As a first step of application,the model shows potential for nucleosome positioning prediction on genome.
出处
《生物物理学报》
CAS
CSCD
北大核心
2010年第5期421-428,共8页
Acta Biophysica Sinica
基金
国家自然科学基金项目(60761001)
内蒙古科技大学创新基金(2009NC066)~~
关键词
核小体定位
多样性增量
支持向量机
Nucleosome positioning
Increment of diversity
Support vector machine
