Method development for quantitative methylation analysis by direct bisulfite sequencing, raw data processing and analysis of the Human Epigenome Project

摘要

Epigenetic deals with flexible biochemical information layers that lie on top of the relatively stable DNA sequence and are involved in control of the structure and functionality of the DNA. One of the most easily analyzed epigenetic layers is DNA methylation.The first part of this work describes the development and assessment of new algorithms enabling methylation quantification by interpretation of sequencing electropherogram data from direct bisulfite sequencing. It is shown that the use of the algorithms on data from PCR products is a suitable replacement for subcloning and sequencing of about 10 subclones - a prerequisite for efficient high throughput DNA methylation studies by direct sequencing, such as carried out in the Human Epigenome Project (HEP).The second part demonstrates the possibility to compensate for artifacts and signal echos in raw data from direct bisulfite sequencing using a deconvolution algorithm.In the third part of this thesis the data of the HEP is analyzed. The HEP is the first large-scale project providing high resolution methylation data in 12 healthy human tissue types on 3 chromosomes analyzed, with a view to answering biological questions. It is shown that differential methylation between healthy tissues is a common phenomenon - especially in conserved non coding sequences, how CpG density and proximity to functional genomic sites influence the methylation profile and in how far CpGs tend to be organized in co-methylated blocks.