gchc: Genetic Correlation Heterogeneity for Causality (GCHC)

In GCHC: Genetic Correlation Heterogeneity for Causality (GCHC)

Description

This function tests the difference between genetic correlation estimates reported by LD Score regression (LDSC), to infer causality.

Usage

gchc(rg1.log = NULL, rg2.log = NULL, exposure.sumstats = NULL,
  outcome.sumstats1 = NULL, outcome.sumstats2 = NULL,
  exposure = "Exposure", outcome = "Outcome", heading = TRUE)

Arguments

rg1.log	LDSC log file for the 1st genetic correlation estimation between the exposure and outcome phenotypes.
rg2.log	LDSC log file for the 2nd genetic correlation estimation between the exposure and outcome phenotypes.
exposure.sumstats	The gzipped file of exposure phenotype GWAS summary statistics, munged by LDSC.
outcome.sumstats1	The gzipped file of outcome phenotype GWAS (1st population) summary statistics, munged by LDSC.
outcome.sumstats2	The gzipped file of outcome phenotype GWAS (2nd population) summary statistics, munged by LDSC.
exposure	Name of the exposure phenotype.
outcome	Name of the outcome phenotype.
heading	logical value that specified whether the software intro heading is printed.

Value

A list is returned with:

• rg.diff The estimated difference between two genetic correlation estimates.

• se The standard error of rg.diff.

• p.value The p-value testing the null hypothesis of rg.diff = 0.

• r.rg The estimated correlation between two genetic correlation estimates.

Note

GWAS of the exposure phenotype is required to be done in only one population, to keep its heritability fixed. Two different populations/sources are required for GWAS of the outcome phenotype.

Author(s)

Xia Shen

References

Shen X, Ning Z, Joshi PK, Lee Y, Wilson JF, Pawitan Y (2017). Genetic correlation heterogeneity detects causal factors for complex traits. Submitted.

See Also

GCHC homepage: http://gchc.shen.se

Examples

## Not run: 
gchc(rg1.log = 'EDU_BMI1.log', 
                rg2.log = 'EDU_BMI2.log', 
                exposure.sumstats = 'EDU.sumstats.gz', 
                outcome.sumstats1 = 'BMI1.sumstats.gz',
                outcome.sumstats2 = 'BMI2.sumstats.gz',
                exposure = 'EA', 
                outcome = 'BMI')

## End(Not run)

GCHC documentation built on May 2, 2019, 5:20 p.m.