It is increasingly recognized that pathway-based analysis, which considers cumulative association between the outcome and a group of SNPs or genes in a biological pathway, can greatly complement the single-SNP approach in understanding genetic determinants of common diseases as well as providing insight into the biological process of complex diseases [10-15]. A pathway-based analysis by Baranzini et al [16] not only confirmed previously identified immunological pathways but also found that neural pathways might be responsible for multiple sclerosis. Joel Hirschhorn [11] pointed out that for many diseases, different risk loci are often clustered in a common pathway, so when a study highlights the role of one or a group of loci in a disease, it also provides important insights and predictive information on the role of other loci within the same biological group. He argued that the primary goal of genome-wide association studies should not be the prediction of individual risk loci but rather the discovery of biological pathways underlying polygenic diseases and traits. The genetic variants revealed in pathway-based analysis could be used to build predictive models for complex diseases, and provide insights on how multiple genetic variants jointly contribute to the etiology of complex human diseases.
