Comprehensive literature review and statistical considerations for GWAS meta-analysis

• Published in: Nucleic acids research Vol. 40; no. 9; pp. 3777 - 3784
• Main Authors: Begum, Ferdouse, Ghosh, Debashis, Tseng, George C., Feingold, Eleanor
• Format: Journal Article
• Language: English
• Published: England Oxford University Press 01.05.2012
• Subjects: Computer programs; Data Interpretation, Statistical; Data processing; Databases, Nucleic Acid; Genome-Wide Association Study; Literature reviews; Meta-Analysis as Topic; Reviews; Software; Statistics; Survey and Summary
• ISSN: 0305-1048; 1362-4962; 1362-4962
• DOI: 10.1093/nar/gkr1255

Over the last decade, genome-wide association studies (GWAS) have become the standard tool for gene discovery in human disease research. While debate continues about how to get the most out of these studies and on occasion about how much value these studies really provide, it is clear that many of the strongest results have come from large-scale mega-consortia and/or meta-analyses that combine data from up to dozens of studies and tens of thousands of subjects. While such analyses are becoming more and more common, statistical methods have lagged somewhat behind. There are good meta-analysis methods available, but even when they are carefully and optimally applied there remain some unresolved statistical issues. This article systematically reviews the GWAS meta-analysis literature, highlighting methodology and software options and reviewing methods that have been used in real studies. We illustrate differences among methods using a case study. We also discuss some of the unresolved issues and potential future directions.