新一代高通量测序Chip-seq数据正规化方法研究

Research on Normalization Method of New Generation High-throughput Sequencing Chip-seq Data

本文针对目前生物信息研究中常见的高通量测序技术Chip-seq数据的正规化问题进行了研究。分析了目前常用的TMR正规化方法和LOWESS正规化方法中没有考虑到基因组的结构对于生物数据分布的影响这一不足,提出了一种新的基于基因组功能注释的LOWESS正规化方法。该方法更符合基因组生物学特征,可以根据基因组本身不同的生物学功能的差异,分区域、分类别进行数据正规化处理,更符合基因组的生物学特征,也具有更高的可靠性。同时可以针对不同研究目的,依据不同的功能区域注释信息有针对性的对该区域进行正规化,具有更高的特异性和灵活性以及更低的时间和空间复杂度。

This paper studies the normalization methods of high - throughput sequencing technology Chip - seq data in cur- rent bioinformatics research. Current normalization methods commonly based TMR or LOWESS did not take into account the impact of structural genomics for the distribution of biological data. Due to this analysis, the paper proposes a new LOW- ESS normalization method based on features of genome annotation. This approach considering the biological characteristics of the genome data can process sub - regional normalization according to the different biological functions of genome itself and has higher reliability. At the same time, the proposed new method could normalize corresponding regions according to the different functional annotation for different research purposes with higher specificity and flexibility, as well as lower time and space complexity.