Dissertation > Excellent graduate degree dissertation topics show

Two-stage Case-control Designs in Genome-wide Association Study

Author: MaZhaoJun
Tutor: ChenFeng
School: Nanjing Medical University
Course: Epidemiology and Biostatistics,
Keywords: Genome-wide association studies Genotyping Single-stage design Two-stage design Case - control design Simulation test
CLC: R341
Type: Master's thesis
Year: 2010
Downloads: 76
Quote: 0
Read: Download Dissertation

Abstract

In recent years, genome-wide association study (GWAS) has become an important means to study genetic variation in complex human diseases . The GWAS concern three main issues are : test performance , false positives , and research funding . The traditional design is a single stage of a case - control design , but to ensure that the test performance of the study , the genotyping funding required more . Limited budget , in order to ensure the performance of the inspection of the study , the study design , therefore , it was suggested that the two-stage and multi- stage case - control design . This study describes the basic principle and implementation steps of the two-stage case - control design , and simulation test discussed in the case of limited research budgets the GWAS in case - control design of test performance (Power), the total type I error rate (FWER), positive false discovery rate (FDR), compared with the single-stage case - control design . Also studied different genetic mode ( addition mode and a dominant mode) , the the different pathogenic allele frequency MAF, genotype relative risk GRR, the different second stage with the first stage , the average for each SNP genotyping costs ratio r under the conditions , the influence on the results . The main conclusions of this study are as follows : (1) funding the same premise , genome-wide association studies , case - control design of the two-stage test performance than single-stage case - control design . (2) the second stage and the first stage of the average SNP genotyping costs higher than r , case - control design of the two-stage test performance decreased. ( 3 ) single- phase case - control design and the two-stage case - control study of the total type Ⅰ error rate FWER than those good about control the default inspection standards before the study 0.05 . Control study using the same test standards total I error rate FWER, case - control design of the two-stage positive false discovery rate FDR always lower than the single-stage case - control design .

Two-stage Case-control Designs in Genome-wide Association Study

Abstract

Related Dissertations