Power for case-control association design
This is a revised version of pbsize which is appropriate for
a case-control design under a range of disease models. Essentially, for given sample
size(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. Note it
provides power estimates for given sample size(s) only.
pbsize2(N,fc=0.5,alpha=0.05,gamma=4.5,p=0.15,kp=0.1,model="additive")
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" |
The returned value is the power for the specified design.
Why is the comparison with power.casectrl so bad?
Jing Hua Zhao
The design follows that of pbsize.
## Not run:
# single calc
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]))
# for a range of sample sizes
pbsize2(p=0.1, N=c(25,50,100,200,500), gamma=1.1, kp=.1, alpha=5e-2, model='r')
# create a power table
f <- function(p)
pbsize2(p=p, N=seq(100,1000,by=100), gamma=1.1, kp=.1, alpha=5e-2, model='recessive')
m <- sapply( X=seq(0.1,0.9, by=0.1), f)
colnames(m) <- seq(0.1,0.9, by=0.1)
rownames(m) <- seq(100,1000,by=100)
print(round(m,2))
library(genetics)
m <- c("multiplicative","recessive","dominant","partialrecessive")
for(i in 1:4) print(power.casectrl(p=0.1, N=50, gamma=1.1, kp=.1, alpha=5e-2,
minh=m[i]))
power.casectrl(p=0.1, N=c(25,50,100,200,500), gamma=1.1, kp=.1, alpha=5e-2,
minh='r')
f <- function(p)
power.casectrl(p=p, N=seq(100,1000,by=100), gamma=1.1, kp=.1, alpha=5e-2,
minh='recessive')
m <- sapply( X=seq(0.1,0.9, by=0.1), f)
colnames(m) <- seq(0.1,0.9, by=0.1)
rownames(m) <- seq(100,1000,by=100)
print(round(m,2))
## End(Not run)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.