gewistLevene: Genome Environment Wide Interaction Search Threshold with...
In WeiAkaneDeng/GEWIST: Gene Environment Wide Interaction Search Threshold

Description Usage Arguments Value Author(s) References Examples

Compute the optimal Variance Prioritization power and the corresponding Levene's test prioritization p-value threshold for a given interaction effect size

1	gewistLevene(p, N, theta_gc, theta_c, M, K = 20000, verbose = FALSE)

`p`	minor allele frequency of the SNP, a number between 0 and 0.5
`N`	sample size
`theta_gc`	proportion of quantitative trait variance explained by the interaction, should be a number between 0 and 1
`theta_c`	proportion of quantitative trait variance explained by the covariate, should be a number between 0 and 1
`M`	total number of SNPs to be tested
`K`	number of procedures, by default, set to be 20,000
`verbose`	logical; if TRUE, function returns a data.frame class object listing the VP power at each p-value, from 0.001 to 1 with 0.001 incremental increase.

A list with three components:

`Optimal_VP_power`	VP power to detect interactions at the optimal Levene's test p-value threshold
`Conventional_power`	power to detect interactions without prioritization, i.e, VP power at Levene's test p-value of 1
`Optimal_pval_threshold`	levene'e test p-value at which optimal VP power is achieved

Wei Q. Deng <dengwq@mcmaster.ca> Guillaume Pare <pareg@mcmaster.ca>

Deng W.Q, Pare G. (2011) A fast algorithm to optimize SNP prioritization for gene-gene and gene-environment interactions. Genetic Epidemiology 35: 729-738. doi: 10.1002/gepi.20624

Pare G, Cook NR, Ridker PM, Chasman DI (2010) On the Use of Variance per Genotype as a Tool to Identify Quantitative Trait Interaction Effects: A Report from the Women's Genome Health Study. PLoS Genet 6(6): e1000981. doi:10.1371/journal.pgen.1000981

Levene H. (1960) Robust tests for equality of variances. In Contributions to Probability and Statistics: Essays in Honor of Harold Hotelling eds:I. Olkin, S.G. Ghurye, W. Hoeffding, W.G. Madow & H.B.Mann, pp.278-292. Stanford: Stanford University Press.

# Given a SNP with minor allele frequency of 10% and a sample 
# of 15,000 individuals, we are interested in testing interactions
# between this SNP and a covariate of effect size #10%. The 
# total number of SNP is 500,000. Assume the interaction 
# explains 0.1% of the quantitative trait variance. 


library(GEWIST)
gewistLevene(p = 0.1, N = 15000, theta_gc =0.1/100, theta_c = 0.1 , M = 500000, 
K = 20000, verbose=FALSE)