Issues in information integration of omics data: microarray meta-analysis for candidate marker and module detection and genotype calling incorporating family information

摘要: Nowadays, more and high-throughput genomic data sets are publicly available; therefore, performing meta-analysis to combine results from independent studies becomes an essential approach increase the statistical power, for example, in detection of differentially expressed genes microarray studies. In addition meta-analysis, researchers also incorporate pathway or clinical information external databases perform integrative analysis. this thesis, I will present three projects which encompass types First, we a comprehensive comparative study evaluate 12 methods simulation real examples by using four quantitative criteria: capability, biological association, stability robustness, propose practical guideline practitioners choose most appropriate method applications. Second, develop meta-clustering construct co-expressed modules 11 major depressive disorder transcriptome datasets, incorporated with GWAS databases. Third, computationally feasible algorithm call genotypes higher accuracy considering family next generation sequencing two purposes: (1) new genotype calling complex families, (2) extend our reference panels analyze family-based sequence small sample size. conclusion, several omics analysis result improves public health significance biomarker biomedical research provides insights help understand underlying disease mechanisms.