SPA.I: Compute rare variants association test scores with and(or)...

In SPAr: Perform rare variants association analysis based on summation of partition approaches

Description

Compute the nonparametric association test scores between a set of rare variants and continous or dichotomous outcomes with and(or) without gene-gene interactions

Usage

SPA.I(x, y, nperm = 100, type = "dichotomous", interaction = 0)

Arguments

x	a matrix of numeric genotypes with each row as an individual and each column as a separate rare SNP. Each genotype should be coded as 0,1 and 2.
y	a vector of phenotypes. The length of y should be the same as the number of rows of x.
nperm	the number of permutations used to obtain the significance level.
type	a string to indicate if the phenoype is continuous or dichotomous. It could take values "dichotomous" or "continuous".
interaction	an integer indicating if the interaction is considered when computing the information score. See Details.

Details

There are three types of interactions: 1 - compute the score without interaction (score I1); 2 - compute the score with interaction (score I2); 0 - compute both scores with and without interaction as well as the adaptive p* score.

Value

I	The numeric value of I score. Could be a vector of 3 values if interaction=0.
pvalue	The p-value of corresponding I score.Could be a vector of 3 values if interaction=0.
nperm	Number of permutations.
call	The matched call.
type	Type of methods used. Could be "I1","I2" or "p*".

Author(s)

Ruixue Fan [e-mail]

References

Fan, R., Lo, S-H (2013). A robust model-free approach for rare variantss association studies incorporating gene-gene and gene-environmental interactions. PLOS ONE 8(12):e83057.

Examples

data(SPA_example)
attach(SPA_example)

## compute p-values of dichotomous traits

pstar_dich <- SPA.I(x,y.d,nperm=10) #compute I1,I2 and p* and their p-values
print(pstar_dich)

I1_dich <- SPA.I(x,y.d,interaction=1,nperm=10) # only compute I1 and its pvalue
print(I1_dich)


## compute p-values of continuous traits

pstar_cont <- SPA.I(x,y.c,nperm=10,type="cont")  #compute I1,I2 and p* and their p-values
print(pstar_cont)

I1_cont <- SPA.I(x,y.c,type="cont",interaction=1,nperm=10) # only compute I1 and its pvalue
print(I1_cont)

SPAr documentation built on May 2, 2019, 7:18 a.m.