Next Generation Sequencing Data Processing And Its Application In Microevolution And Cancer Metabolism

With the development of next-generation sequencing(NGS), a huge number of omics data emerging. In recent years, integrated use of these omics data gradually become research hotspots and difficulties. Integration analysis of different levels of omics data are very necessary for research complicated biological scientific problems. In this paper, we reviewed the principle of sequencing technology and its development process, given the processing flow and utility software for NGS data. My research mainly focuses on the integration and application of NGS data and other omics data. The following two relatively independent works were carried out:1. Research on the relationships between nucleosome positioning evolution and gene expression noise.Gene expression noise provides a large number of genetic resources for the adaptive evolution of organisms. Currently, the gene expression variation between individuals cannot be fully explained by the genome mutation. The epigenetic variation was considered closed related to this microscopic gene expression noise. Nucleosomes are the basic units of chromatin, and they are hotspots in epigenetic research field. Nucleosomes spread on the genomes dynamically, especially the promoter region of genes. There were lots of studies reported that the organization of nucleosomes were play a vital role in gene expression variation. However, the mechanism of regulation of gene expression noise was still unknown. Here, we investigated the relationships between nucleosome positioning dynamic evolution and the gene expression noise. We comparative analysis the nucleosome positioning and gene expression of a parental sample and five offspring samples which were paralled mitotic cultivate about 1740 generation from this parental sample. We found there were about 25 bps shift between parental and offspring samples. In addition, nucleosome variations were mainly affect genes which were lower expression, so as to reduce gene expression noise at whole genome scale. Our work helps to understand the regulation of gene expression noise, and provide a new insight into the evolutionary process of cells from lower to higher organisms.2. Personalized quantitative metabolic modeling of liver cancer by using TCGA and HPA omics data.The liver is the largest metabolic organ in the human body, and in charge of a variety of synthesis and decomposition metabolic reactions. The hepatocellular carcinoma is one of cancer which has the highest rate of morbidity and mortality, especially for China, the morbidity and mortality rate are far more than the world average level. In recent years, with the development of genome wide metabolic network reconstruction methods, we have developed a Personalized Quantitative Metabolic Modeling (PQMM) method which could reconstruction personalized quantitative model of metabolic networks. This method was useful for the simulation of organic specific metabolic network, and it broke through the bottlenecks that the metabolic network cannot be constructed to perform with quantitative simulation. Based on the PQMM method, we integrated the TCGA expression data and HPA proteomics data to reconstruct the quantitative modeling of Hepatocellular Carcinoma (HCC) metabolic networks. The results reveal HCC growth rapidly because it’s changed a huge number of cell metabolic reactions. The metabolic reactions which were necessary for normal liver function were significantly reduced. However, cell growth related pathways in HCC cells were upregulated, for example, the pentose phosphate pathway and amino acid synthesis. Besides, we found the ATP production in HCC cells were upregulated by improving the glutamine catabolism rate. Furthermore, we divided the HCC samples into two classes by performing the unsupervised clustering of the metabolic reaction flux. The survival curve of these two samples of HCC was significant distinction with each other, it showed the more ATP production the prognosis is worse, the better prognosis class samples retained part of the normal metabolism of liver function. Our study is the first personalized quantitative metabolic modeling of HCC and normal livers, the results of simulation are highly consistent with experimental data. Our results have important theoretical and significance for prevention and targeted therapy of liver cancer in clinical trials.