Description Usage Arguments Details Value Author(s) References Examples
View source: R/twoStageGwasPower.R
Computes the critical values for a two-stage gwas association study and compute power for replication and joint analyses.
1 | twoStageGwasPower(pD, pG, grr, inheritance = "multiplicative", pi.samples, pi.markers, alpha.marker, n.cases, n.controls)
|
pD |
Probability of disease in the population (prevalence) |
pG |
Frequency of disease allele in the population |
grr |
Genotypic relative risk |
inheritance |
Inheritance type ("dominant", "recessive", "multiplicative", or "additive" |
pi.samples |
Proportion of samples genotyped in Stage 1 |
pi.markers |
Proportion of markers genotyped in Stage 2 |
alpha.marker |
Significance level used for each marker, accounting for multiple comparisons among a large number of markers |
n.cases |
Number of cases |
n.controls |
Number of controls |
This function computes the critical values and powers of the replication and joint methods of analyzing a two-stage GWAS design. Details may be found in Skol AD, Scott, LJ, Abecasis GR, Boehnke M (2006)
A list containing:
power.SingleStage |
Power of a one stage design |
power.joint |
Power of a joint analysis |
power.rep |
Power of a replication analysis (based only on the second stage markers) |
c1 |
Stage one threshold |
c2 |
Replication (stage two) threshold |
c.joint |
Joint analysis threshold |
c.singleStage |
Single stage design threshold |
penetrance.GG |
Penetrance of the GG genotype (homozygous for disease allele) |
penetrance.Gg |
Penetrance of the Gg genotype |
penetrance.gg |
Penetrance of the gg genotype |
p0 |
disease allele frequency in controls |
p1 |
disease allele frequency in cases |
p.stageOne |
probability that associated markers will be followed up in Stage 2 |
savings |
reduction in genotyping using two-stage design as compared to the single-stage design |
Dirk F Moore <dirkfmoore@gmail.com>
Skol AD, Scott, LJ, Abecasis GR, Boehnke M (2006) Nature Genetics doi:10.1038/ng1706
http://www.sph.umich.edu/csg/abecasis/CaTS/
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 | # prevalence of disease is 0.10, the allele frequency is 0.40,
# a multiplicative model with 0.40 samples in the first stage and
# 10% of the markers selected for Stage 2. There are 1000 cases
# and 1000 controls, 300,000 markers, with a genome-wide alpha of 0.05
power.gwas.out <- twoStageGwasPower(pD=0.10, pG=0.40, grr=1.40,
inheritance="multiplicative", pi.samples=0.40, pi.markers=0.10,
alpha.marker=0.05/300000, n.cases=1000, n.controls=1000)
power.gwas.out
# Same, but with 1% of markers selected for Stage 2
power.gwas.out2 <- twoStageGwasPower(pD=0.10, pG=0.40, grr=1.40,
inheritance="multiplicative", pi.samples=0.40, pi.markers=0.010,
alpha.marker=0.05/300000, n.cases=1000, n.controls=1000)
power.gwas.out2
# Same, but a dominant model with 4000 controls and 2000 cases
power.gwas.out3 <- twoStageGwasPower(pD=0.10, pG=0.40, grr=1.40,
inheritance="dominant", pi.samples=0.40, pi.markers=0.10,
alpha.marker=0.05/300000, n.cases=2000, n.controls=4000)
power.gwas.out3
|
Power for one-stage design 0.7437531
Power for replication analysis 0.4561015
Power for joint analysis 0.7419231
Threshold for one-stage design 5.233126
Threshold for stage one 1.644854
Threshold for replication design 4.649133
Threshold for joint design 5.230999
Penetrance for GG 0.1456599
Penetrance for Gg 0.1040428
Penetrance for gg 0.07431629
Disease allele frequencies in cases 0.4827586
Disease allele frequencies in controls 0.3908046
Probability that associated markers will be followed up in stage two 0.9811959
Reduction in genotyping from two-stage design 0.54
Power for one-stage design 0.7437531
Power for replication analysis 0.5768826
Power for joint analysis 0.7110829
Threshold for one-stage design 5.233126
Threshold for stage one 2.575829
Threshold for replication design 4.14941
Threshold for joint design 5.204231
Penetrance for GG 0.1456599
Penetrance for Gg 0.1040428
Penetrance for gg 0.07431629
Disease allele frequencies in cases 0.4827586
Disease allele frequencies in controls 0.3908046
Probability that associated markers will be followed up in stage two 0.8745411
Reduction in genotyping from two-stage design 0.594
Power for one-stage design 0.5358034
Power for replication analysis 0.2867125
Power for joint analysis 0.5341269
Threshold for one-stage design 5.233126
Threshold for stage one 1.644854
Threshold for replication design 4.649133
Threshold for joint design 5.230999
Penetrance for GG 0.111465
Penetrance for Gg 0.111465
Penetrance for gg 0.07961783
Disease allele frequencies in cases 0.4458599
Disease allele frequencies in controls 0.3949045
Probability that associated markers will be followed up in stage two 0.9574386
Reduction in genotyping from two-stage design 0.54
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.