MAGENTA: Meta-Analysis Gene-set Enrichment of variaNT Associations

Image credit: Lauren Solomon, Broad Communications, Broad Institute, Cambridge, MA.

MAGENTA is a computational tool that tests for enrichment of genetic associations in predefined biological processes or sets of functionally related genes, using genome-wide genetic data as input. MAGENTA is designed to analyze datasets for which genotype data are not readily available, such as large genome-wide association study (GWAS) meta-analyses. It can be used either (i) to test a specific hypothesis or (ii) to generate hypotheses by testing a range of known bioogical gene sets (we provide pathways from several public databases).

The only input required is a table with variant association p-values and their chromosome positions taken from a genome-wide association study or meta-analysis. Optional: pathway/s or gene set/s of interest. Otherwise, we provide a set of pathways from public databases (see below).

The main output of MAGENTA is a nominal gene set enrichment analysis (GSEA) p-value and a false discovery rate for each gene set or pathway tested. There are several options, including running MAGENTA in the absence of a subset of genes, such as a predefined set of disease or trait genes. Additional information is provided, such as the expected and observed number of genes above the enrichment cutoff, and the number and name of genes in each tested gene set that lie near validated disease or trait SNPs if inputed by the user.

Another feature that MAGENTA provides is a print out of gene association p-values, corrected for SNP to gene score confounders, for all genes in pathways or gene sets of interest, as well as other gene-cetric properties including the most-significant SNP p-value per gene region and its original GWAS p-value.

MAGENTA was developed by Ayellet Segrè in the labs of David Altshuler and Mark Daly at the Broad Institute, the Center for Human Genetic Research of Massachusetts General Hospital, and Harvard Medical School. MAGENTA is described in the reference below.


MAGENTA version 2.4, July 2011 can be downloaded here for local use. New addition: MAGENTA can now run on genome build 37 (hg19) or build 36 (hg18). Instructions in README file; Software is written in Matlab and requires Statistical toolbox; Biological gene sets from KEGG, PANTHER, Reactome, BioCarta and Gene Ontology are available.


    Ayellet V. Segrè, DIAGRAM Consortium, MAGIC investigators, Leif Groop, Vamsi K. Mootha, Mark J. Daly, and David Altshuler (2010). Common Inherited Variation in Mitochondrial Genes is not Enriched for Associations with Type 2 Diabetes or Related Glycemic Traits. PLoS Genetics Aug 12;6(8). pii: e1001058.


    For questions or comments please contact A S E G R E at broadinstitute dot org.

    Broad Logo MGH Logo HMS Logo