Abstract
Homocysteine is a risk factor for atherosclerosis, and the level of homocysteine in plasma is known to be strongly influenced by genetic factors-not only rare variants, but also common polymorphisms. This report describes a comprehensive postgenomic strategy for elucidating useful genetic information about homocysteine metabolism. The standard method for gathering such information is the candidate gene approach, which is an effective method based on known biological information. After collecting evidence from independent research projects, a critical epidemiological review permits a determination as to whether a putative association is true or not. A genome-wide association study (GWAS), which requires no biological information, can identify new candidates and confirm associations suggested by the candidate gene approach. The importance of methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism, which was shown in a randomized controlled trial conducted by the present author, and in other studies, was independently confirmed by a large-scale GWAS. GWASs have also identified new candidate genes, but these must be confirmed by independent studies. In homocysteine metabolism, the classical candidate gene approach was sufficiently robust to detect the true association. However, candidate markers newly discovered by GWAS need to be confirmed by well-designed epidemiological studies to determine their significance. International statements, such as CONSORT and STREGA, provide useful principles for conducting such research.