Nucleotide-Level Variant Analysis of Next-Generation Sequencing Data Using a Cloud-Based Data Analysis Pipeline

摘要

To demonstrate the flexibility of a cloud-based solution for analyzing disparate sets of next-generation sequencing data, we looked at carefully chosen samples across different populations from the 1,000 Genomes Project ( www.1000genomes.org ) and conducted an extensive analysis on two Chinese populations, the “Chinese in Beijing” (CHB) and the “Chinese in metropolitan Denver” (CHD), each consisting of 28 exomes. Each dataset was uploaded into the system using raw data files acquired from the 1,000 Genomes Project. Using these data and a cloud-based data analysis pipeline, we performed a nucleotide-level variant analysis combined with a population allele frequency analysis across all samples for the two populations. To identify alleles that are significantly different across the two populations, a Pearson's chi-square test was applied, which resulted in a total of 1.5 Mio SNPs, of which 84 were non-synonymous with a p-value of less than 0.01. Interestingly, the genes associated with non-synonymous variants of the Chinese in metropolitan Denver population were enriched for biological annotations such as endocrine system disorder, metabolic disease, cardiac fibrosis, and inflammation (includes ZNF264, RPS6KA2, ROBO2, CRK, MUSK, CBL, CRK, and others). Furthermore, genes usually associated with liver injury were also identified for this population, suggesting the liver is exposed to toxic agents more so in this population compared to the CHB population. The observed genomic differences in these two different Chinese populations living in different parts of the world hint towards a potential link between nutrition and different diseases (e.g. heart disease or metabolic diseases). Using this analysis as a case study, we will demonstrate how a scalable computational infrastructure can provide researchers and sequencing service providers alike, a cost effective and secure cloud-based computing platform as a powerful and collaborative technology solution for large scale sequence data analysis and management.