trio.power: Power and sample size calculation

In trio: Testing of SNPs and SNP Interactions in Case-Parent Trio Studies

Description

Computes power for genotypic TDT, allelic TDT or Score test given n trios or required sample size to gain given power.

Usage

trio.power(maf = 0.5, RR = 1.5, alpha = 5*10^(-8), n = NULL, beta = NULL, 
  model = c("additive", "dominant", "recessive"), test = c("gTDT", "Score", "aTDT"))
  
## S3 method for class 'trio.power'
print(x,digits=4,...)

Arguments

maf	a numeric vector of population frequencies of a mutant allele.
RR	a numeric vector of the assumed relative risks for an individual getting a disease with 1 (in case of recessive model 2) mutant alleles compared to the risk of individuals carnying 0 mutant alleles.
alpha	a numeric vector of significance levels (Type I Error probability).
n	a numeric vector containing number of trios in a study. Must be filled for power calculation. Must not be NULL for sample size calculation.
beta	the desired power of the test. Must be filled for power calculation. Must not be NULL for sample size calculation.
model	a character containing the genotypic model assumed. Possible values are "additive", "dominant" and "recessive". In case of test="aTDT", the standard multiplicative model will be considered. Abbreviations are allowed. Thus, e.g., model = "dom" will fit a dominant model, and model = "r" a recessive model.
test	the chosen test. Must be "aTDT", "gTDT" or "Score". Abbreviations are allowed. Thus, e.g., test = "g" will perform a genotypic TDT, and test = "S" a Score test.
x	an object of class trio.power.
digits	number of digits that should be printed.
...	ignored

Details

Power and sample size calculation is derived on Knapp (1999). The power or the sample size will be calculated for all combinations of p, RR, alpha, test, model and n or beta.

Value

An object of class trio.power containing the following numeric values or vectors, respectively:

model	the chosen model
size	In case of sample size calculation: calculated sample sizes
beta	In case of sample size calculation: desired power
n	In case of power calculation: given number of trios
power	In case of power calculation: calculated power
alpha	Type I error
test	the chosen test
RR	the relative risks assumed
p	the assumed allele frequency
calc	the type of calculation

Author(s)

Christoph Neumann

References

Knapp, M. (1999). A Note on Power Approximations for the Transmission/Disequilibrium Test. American Journal of Human Genetics, 64, 1177-1185.

Neumann, C., Taub, M.A., Younkin, S.G., Beaty, T.H., Ruczinski, I., Schwender, H. (2014). Analytic Power and Sample Size Calculation for the Genotypic Transmission/Disequilibrium Test in Case-Parent Trio Studies. Submitted.

Examples

# The required samples size to reach of power
# of 0.8 when testing SNPs with minor allele
# frequencies of 0.1 and 0.2 with an additive
# or dominant genotypic TDT and score test
# can be determined by 
trio.power(maf = c(0.1, 0.2), beta = 0.8, model = c("add", "dom"))

Example output

       Trio studies sample size calculation 

    Test          Model MAF alpha  RR beta Trios
1   gTDT       additive 0.1 5e-08 1.5  0.8  2524
2   gTDT       additive 0.2 5e-08 1.5  0.8  1607
3   gTDT       dominant 0.1 5e-08 1.5  0.8  2771
4   gTDT       dominant 0.2 5e-08 1.5  0.8  1950
5  Score       additive 0.1 5e-08 1.5  0.8  2505
6  Score       additive 0.2 5e-08 1.5  0.8  1596
7  Score       dominant 0.1 5e-08 1.5  0.8  2749
8  Score       dominant 0.2 5e-08 1.5  0.8  1935
9   aTDT multiplicative 0.1 5e-08 1.5  0.8  2505
10  aTDT multiplicative 0.2 5e-08 1.5  0.8  1596

trio documentation built on Nov. 8, 2020, 7:41 p.m.