/babelomics

Burden test

Francisco García edited this page Jan 30, 2015 · 9 revisions

General

Tutorial

Analysis tools

Worked examples

Clone this wiki locally

Although whole-genome association studies using SNPs are a powerful approach for detecting common variants, they are underpowered for detecting associations with rare variants.

Some methods used for analysis of common variants are applicable to sequence data but their performance might not be optimal. Collapsing methods, which involves collapsing genotypes across variants and applying a univariate test, are powerful for analyzing rare variants, whereas multivariate analyses are robust against inclusion of noncausal variants. Both methods are superior to analyzing each variant individually with univariate tests.

For association testing with rare variants, it is customary to aggregate information across several variant sites within a gene to enrich association signals and to reduce the penalty of multiple testing. The simplest approach is the burden test, which creates a burden score for each subject (by taking a weighted linear combination of the mutation counts within a gene or indicating whether there is any mutation within a gene)

In order to unify the advantages of both collapsing and multiple-marker tests, we have implemented in Babelomics a Combined Multivariate and Collapsing (CMC) method using AssotesteR package.

The CMC method is a pooling approach proposed by Li and Leal (2008) that uses allele frequencies to determine the partition of the variants into groups. After the rare variants are selected, they are collapsed into an indicator variable, and then a multivariate test such as Hotelling’s T2 test is applied to the collection formed by the common variants and the collapsed super-variant.

References

Find the Babelomics suite at http://babelomics.org