FORGE Basics

[[Home|Home]]

Genome-Wide Association Studies (GWAS) have proven to be a powerful tool to shed light on the genetic basis of many human traits. See the [[GWAS Catalogue|www.genome.gov/26525384]] for an updated collection of GWAS results. Traditional GWAS analysis undergoes the analysis of each genetic variant individually, which can involve several millions of univariate tests. Is has become clearer that many studies are unpowered to detect, at genome-wide significant level, low effect sizes and that many of these weak effect may underlie common disorders. Aiming to enhance power of GWAS we have developed FORGE. FORGE allows to combine all associations signals in a genomic region, e.g. a gene or biological pathway. Interestingly, we have found that this approach improves the power of GWAS to find true disease genes, meaning those that are confirmed by replication in larger studies. We also found that FORGE provides similar results to those found by haplotypic analysis. The gene p-values are accurate and in our manuscript we showed that attain high correlation with empirical gene p-values calculated by shuffling phenotype labels. Overall, FORGE provides asymptotic p-values that accurate and feasible to calculate in modest computers (3-4 hours with 1 CPU and 4 GB of RAM). We are using FORGE gene p-values in pathway or network analyses and found it very good for that. Please check our references in the page for new publications using FORGE.

FORGE is implemented in Perl and should be platform independent.

We have also written tutorials to install and run FORGE and additional tutorials to perform pathway and network analysis with FORGE GWAS results. Having learn this it should be easy to add FORGE to you GWAS analysis pipelines.

For those who prefer not to deal with the command line we provide the ForgeWeb Server. You only need to upload your GWAS results and we will send you the gene p-values by e-mail. Check the [[ForgeWeb server|https://compbio.brc.iop.kcl.ac.uk:8443/forgeweb]].

[[Home|Home]]