| ld.design {ldDesign} | R Documentation |
Find the sample size required to detect linkage disequilibrium with a given Bayes factor, with a given power, or find the power of experimental designs to detect linkage equilibrium with a given Bayes factor.
ld.design(p, q, D, h2, phi, Bf, power, nmin = 50, nmax = 1e+05, ninterp = 50,
missclass.rate = 0, print.it = FALSE)
ld.power(n, p, q, D, h2, phi, Bf, missclass.rate = 0)
n |
ld.power: vector of sample sizes |
p |
Bi-allelic marker allele frequency |
q |
Bi-allelic QTL allele frequency |
D |
Linkage disequilibrium coefficient |
h2 |
QTL `heritability', i.e. proportion of total or phenotypic variance explained by the QTL |
phi |
Dominance ratio: phi = 0 denotes purely additive,
phi = 1 denotes purely dominant allele effects |
Bf |
Bayes factor |
power |
ld.design: Power, or probability of detecting an effect with Bayes
factor greater than Bf |
nmin |
ld.design: Lower bound for sample size |
nmax |
ld.design: Upper bound for sample size |
ninterp |
ld.design: Number of sample sizes to try |
missclass.rate |
Proportion of marker values which are missclassified, i.e. incorrect (to allow for genotyping errors) |
print.it |
If TRUE print results for sample sizes tried |
These functions implement the method described in Ball (2003) for
obtaining the power of designs for detecting linkage disequilibrium with
a given Bayes factor. The F values, (and hence significance levels)
corresponding to the given Bayes factors, sample sizes, and marker genotype
frequecies, are calculated using the method of Spiegelhalter and Smith (1982)
(R functions oneway.bf.alpha.required,
SS.oneway.bf). The power is obtained using a corrected
version of the classical deterministic power calculation from
Luo (1988) (R function luo.ld.power).
For ld.power, a matrix with columns:
n |
Sample sizes |
power |
Power of the design with the given sample sizes |
Additionally the return value has attributes indicating the
linkage disequilibrium parameters used.
For ld.design the sample size is returned.
Rod Ball rod.ball@forestresearch.co.nz www.forestresearch.co.nz
Ball, R.D. 2003 Experimental designs for reliable detection of linkage disequilibrium in unstructured random population association studies.
Luo, Z.W. 1988 Detecting linkage disequilibrium between a polymorphic marker locus and a trait locus in natural populations. Heredity 80, 198–208
Spiegelhalter, D. and A.F.M. Smith 1982 Bayes factors for linear and log-linear models with vague prior information J. Royal Statist Soc. B 44: 377–387.
luo.ld.power, ld.sim, oneway.bf.alpha,
oneway.bf.alpha.required, SS.oneway.bf
ld.power(n=seq(100,1000,by=100),p=0.5,q=0.5,D=0.1,h2=0.1,phi=0,Bf=20) ld.design(p=0.5,q=0.5,D=0.1,h2=0.1,phi=0,Bf=20,power=0.9,print.it=TRUE,nmin=600,nmax=4000) ld.design(p=0.5,q=0.5,D=0.1,h2=0.1,phi=0,Bf=20,power=0.9,print.it=FALSE,nmin=1700,nmax=1900)