An Experiment to Whole-genome Resequencing Data Analysis (III)

(5). Functional analysis and annotation of variation

(A). Functional annotation of SNP, Indels and large structural variation SV;

(B). The SNPs on the coding region of the transcript containing the translation start annotation information are classified as: synonymous SNPs, non-synonymous SNPs and nonsense SNPs (introduced terminators), SNPs that interfere with terminators, and interference with splice junctions Point SNP; in order to reduce false positives, we use strict screening methods to filter errors from indels;

(C). Functional analysis of mutations in the missense coding region: Informatics analysis algorithms are used to evaluate the effect of somatic mutations relative to germline mutations on protein structure and function.

(6). Association analysis between SNV, SNP and miRNA research

miRNA is a small molecule that plays an important regulatory role. We will analyze the pri-mRNA, pre-miRNA and miRNA target gene sequences of miRNA to identify potential SNP functional sites. According to the literature research, evidence is provided that there are SNPs in different positions in the secondary structure of Human pre-miRNA. We will evaluate the effect of SNP on the structure of pre-miRNA through thermodynamic stability analysis; in addition, we will also target miRNA-Target Gene interaction sites were analyzed to assess the impact of SNP on target gene targeting.

(7). Association analysis between SNV, SNP and GWAS research

Analyze the distribution of OR values ​​of susceptible genes obtained in the GWAS study at different coordinates on the genome; compare the currently known GWAS research results with SNP; in-depth discussion of the relationship between SNP and susceptible genes based on LD linkage imbalance The direct and indirect association methods can separately identify SNPs related to phenotype. For SNPs that are not easy to obtain and locate, the mutation loci of disease susceptibility genes can be inferred through LD linkage imbalance.

(8) Analysis of biological pathways (metabolic pathways, signaling pathways)

Biological pathways, including metabolic pathways and signal transduction pathways, are important components of biological functions. We put the corresponding genes of various forms of mutations and mutations, including SNV and SNP, into the biological pathways for synthesis. Analyze and investigate the degree of influence and the law of influence of functional mutations on pathways. Through the GSEA (with chip expression profiling data), KS test, hypergeometric distribution test and other methods to sort the enrichment of variant genes in certain pathways, identify potential pathways for functional changes.

(9). Protein-protein interaction (PPI) network analysis

Protein interaction is also an important way for the gain and loss of biomolecule functions. Therefore, we systematically analyze the mutant proteins in the protein interaction network and the affected network node proteins, and perform functions on the affected network substructure. Annotation analysis and cluster rich analysis. We use a network analysis algorithm to analyze the functional enrichment of subnetworks affected by various mutations;

(10). Analysis of cis-gene regulation network module (CRM)

(a) Promoter sequence analysis

Include Motif prediction on the mover region, and compare with the TFBS binding sites in the database of known transcription factors TRANSFAC and JASPAR;

Conservative analysis on the promoter region to analyze the relationship between mutation positions and conserved regions;

(b) Calculate genome-wide conservation. Determine the conservation of TFBS and the conservation of mutation position;

(11) Whole-genome statistics of rearrangements and mutations

(A). Arrangements of somatic cells and germline

Somatic mutations are a type of content that requires important analysis relative to germ line mutations. Our sequencing scheme designed for Case-control can analyze mutations separately, including SNV, indel, and CNV; if only in tumor / disease (Case Group) does not appear in the normal (control group) mutations we can be considered somatic somatic mutations. By comparing the somatic mutation with the dbSNP database, you can find potential new mutations and recorded mutation positions. Then, the mutations are compared to the gene region and the non-gene region, respectively. The gene region specifically includes: intron region, UTR, splice site region and exon region. Among them, the exon statistics are respectively: synonymous, deletion, frameshift, insertion, missense, nonsense and noncoding protein exons (Non-protein coding exon) and other different types. Synthesizing the analysis results of different aspects, and according to the mutation classification, it gives the statistical data table of the number of types of arrangements: SNV and CNV. Label each sequencing sample separately, including somatic mutations and germline mutations.

(B) Analysis of global rearrangement distribution characteristics of whole genome

Mainly (a) structural variation between and within chromosomes, (b) heterozygosity (LOH) and allelic imbalance, (c) copy number variation (gain or deletion) and high-confidence SNV (Statistics in the interval of 1Mb) and other different conditions are displayed separately on different levels of the ring diagram in accordance with the karyotype

(C) Single nucleotide mutation trend and pattern analysis

Count the main forms of transversion and their respective proportions at the level of somatic cells and germline cells (a); if there is expression profile data, you can analyze the number or type of mutation rearrangements of expressed genes and non-expressed genes ( b); Somatic variation in the upstream region of the transcription initiation site, germline germline variation and random variation of the respective number statistics (c) and the known 210 kinds of different tumor disease mutation spectrum for comparison

1. Natural selection analysis

The somatic mutations we observed through sequencing may have been caused by a complicated process. Therefore, we need to do in-depth analysis in studying the origin of these mutations, how mutations are affected by DNA repair mechanisms, and the laws of mutations in the development and evolution of diseases. Natural selection generally plays a role in two aspects, that is, to retain mutations that are conducive to disease development and evolution, while restricting mutations in important functional regions of the genome, such as transcriptional regulatory regions and protein-coding regions. Therefore, (1) If the experimental design is to compare primary disease with normal control, systematic analysis can analyze the possible mechanisms and natural selection factors of complex diseases in the process of forming mutations. (2) If the experimental design is based on samples of lesions and their metastasis or adjacent locations, we can construct a model of mutation evolution and metastasis to analyze the dynamic mode of mutation and the pattern of unstable state mutations in the genome.

Judgment of positive selection: Analyze the positive selection trends of SNP and SNV regions, and explain the functionality of SNV and SNP at the level of evolution and population genetics; for control and case group samples, we use different statistical algorithms to calculate SNP, CNV in each The difference in the sample, and then find the SV with positive selection features.