CJAMP (version 0.1.1)

compute_expl_var: Phenotypic variance explained by genetic variants.

Description

Function to estimate the percentage of the variance of a phenotype that can be explained by given single nucleotide variants (SNVs).

Usage

compute_expl_var(genodata = NULL, phenodata = NULL,
  type = "Rsquared_unadj", causal_idx = NULL, effect_causal = NULL)

Arguments

genodata

Numeric vector or dataframe containing the genetic variant(s) in columns. Must be in allelic coding 0, 1, 2.

phenodata

Numeric vector or dataframe of the phenotype.

type

String (vector) specifying the estimation approach(es) that are computed. Available are the methods "Rsquared_unadj", "Rsquared_adj", "MAF_based", and "MAF_based_Y_adjusted". See below for more details.

causal_idx

Vector with entries TRUE, FALSE specifying which SNVs are causal. Has to be supplied for the approaches MAF_based and MAF_based_Y_adjusted.

effect_causal

Numeric vector containing the effect sizes of the causal SNVs. Has to be supplied for the approaches MAF_based and MAF_based_Y_adjusted.

Value

A list containing the estimated percentage of explained phenotypic variance.

Details

Four different approaches are available to estimate the percentage of explained phenotypic variance (Laird & Lange, 2011):

(1) "Rsquared_unadj": Unadjusted \(R^2\) from a linear regression of the phenotype conditional on all provided SNVs.

(2) "Rsquared_adj": Adjusted \(R^2\) from a linear regression of the phenotype conditional on all provided SNVs.

(3) "MAF_based": Expected explained phenotypic variance computed based on the MAF and effect size of the provided causal SNVs.

(4) "MAF_based_Y_adjusted": Expected explained phenotypic variance computed based on the MAF and effect size of the causal SNVs, with respect to the empirical phenotypic variance, which is the broad-sense heritability relative to the empirical phenotypic variance.

References:

Laird NM, Lange C (2011). The fundamentals of modern statistical genetics. New York: Springer.

Examples

# NOT RUN {
set.seed(10)
genodata <- generate_genodata(n_SNV = 20, n_ind = 1000)
phenodata <- generate_phenodata_1_simple(genodata = genodata[,1],
                                         type = "quantitative", b = 0)
compute_expl_var(genodata = genodata, phenodata = phenodata$Y,
                 type = c("Rsquared_unadj", "Rsquared_adj"),
                 causal_idx = NULL, effect_causal = NULL)

# }