pbsize2: Power for case-control association design

Description

Power for case-control association design

Usage

pbsize2(
  N,
  fc = 0.5,
  alpha = 0.05,
  gamma = 4.5,
  p = 0.15,
  kp = 0.1,
  model = "additive"
)

Value

The returned value is the power for the specified design.

Arguments

N: The sample size.
fc: The proportion of cases in the sample.
alpha: Type I error rate.
gamma: The genotype relative risk (GRR).
p: Frequency of the risk allele.
kp: The prevalence of disease in the population.
model: Disease model, i.e., "multiplicative","additive","dominant","recessive","overdominant".

Details

This extends pbsize from a multiplicative model for a case-control design under a range of disease models. Essentially, for given sample sizes(s), a proportion of which (fc) being cases, the function calculates power estimate for a given type I error (alpha), genotype relative risk (gamma), frequency of the risk allele (p), the prevalence of disease in the population (kp) and optionally a disease model (model). A major difference would be the consideration of case/control ascertainment in pbsize.

Internally, the function obtains a baseline risk to make the disease model consistent with Kp as in tscc and should produce accurate power estimate. It provides power estimates for given sample size(s) only.

Examples

Run this code

if (FALSE) {
# single calculation
m <- c("multiplicative","recessive","dominant","additive","overdominant")
for(i in 1:5) print(pbsize2(N=50,alpha=5e-2,gamma=1.1,p=0.1,kp=0.1, model=m[i]))

# a range of sample sizes
pbsize2(p=0.1, N=c(25,50,100,200,500), gamma=1.2, kp=.1, alpha=5e-2, model='r')
  
# a power table
m <- sapply(seq(0.1,0.9, by=0.1),
            function(x) pbsize2(p=x, N=seq(100,1000,by=100),
                        gamma=1.2, kp=.1, alpha=5e-2, model='recessive'))
colnames(m) <- seq(0.1,0.9, by=0.1)
rownames(m) <- seq(100,1000,by=100)
print(round(m,2))
}