Genome
wide association study (GWAS) have revolutionised the field of gene mapping. GWASs
typically calculate a statistical association of each SNP to the trait,
considering each as an independent entity. Although the single SNP approach has
helped to find many crucial variants associated with traits, the overall effect
of all loci found associated with particular trait typically accounts for only
a small proportion of its heritability. To help explain this missing
heritability, we propose a versatile pathway analysis method for GWAS (Pathway-VEGAS).
The method is based on prior calculation of gene-based p-values using the
existing Versatile Gene-based Association Study (VEGAS) software. Pathway-VEGAS
uses the gene-based P-values to construct a test for any pathway of interest,
typically using a set of pre-specified pathways we have assembled. The method
appropriately takes into account cases where neighbouring genes are present in
the same pathway - new p-values for relevant regions are calculated by accounting
for linkage disequilibrium between markers using simulations from the
multivariate normal distribution. Pathway size is taken into account via a re-sampling
approach. This method can be applied in all SNP association studies, including
Meta-analysis, Case-control studies, Family based GWAS, DNA-Pooling based GWAS
and Singleton data.We found
biologically significant findings while applying the method on GWAS outcome for
a number of traits including Endometriosis.