Skip to content

sanderslab/wgsPowerCalc

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

wgsPowerTest

This R package runs power calculations for the discovery of variants in whole genome sequencing data

Install the wgsPowerCalc project in R:

devtools::install_github("stephansanders/wgsPowerTest")
library(wgsPowerCalc)

Set the working directory for where you want the PDFs to be printed

setwd("/path/to/directory/")

Run an example to estimate the power to detect an excess of de novo protein truncating variants (PTVs) in autism spectrum disorder as the relative risk is varied:

  • Maximum relative risk (R) = 25
  • Sample size (N) = 5000 cases
  • Case to control ratio (r) = 1
  • P-value threshold (p_thres_denovo_burden) = 0.05
  • Number of de novo PTVs per person (q) = 0.0996
  • Proportion of de novo PTVs that mediate risk (f) = 0.05
  • Prefix for PDF of power (name) = Test1
  • Color of line in PDF (col) = "#4daf4a"
plotDnBurdenByRelativeRisk(R=25, N=5000, r=1, q=0.0996, f=0.05, p_thres_denovo_burden=0.05, name="Test1", col="#4daf4a") 

Run an example to estimate the power to detect an excess of rare protein truncating variants (PTVs) in autism spectrum disorder as the relative risk is varied:

  • Maximum relative risk (R) = 2.5
  • Sample size (N) = 20000 cases
  • Case to control ratio (r) = 1
  • P-value threshold (p_thres_cc_burden) = 0.05
  • Number of rare PTVs per person (q) = 5.04
  • Proportion of rare PTVs that mediate risk (f) = 0.05
  • Prefix for PDF of power (name) = Test2
  • Color of line in PDF (col) = "#4daf4a"
plotCcBurdenByRelativeRisk(R=2.5, N=20000, r=1, q=5.04, f=0.05, p_thres_cc_burden=0.05, name="Test2", col="#4daf4a")

Run an example to estimate the power to detect a specific locus through an excess of de novo protein truncating variants (PTVs) in autism spectrum disorder as the relative risk is varied:

  • Maximum relative risk (R) = 25
  • Sample size (N) = 5000 cases
  • Case to control ratio (r) = 1
  • P-value threshold (p_thres_denovo_locus) = 2.5e-06
  • Number of de novo PTVs per person (q) = 0.0996
  • Proportion of de novo PTVs that mediate risk (f) = 0.05
  • Prefix for PDF of power (name) = Test3
  • Color of line in PDF (col) = "#4daf4a"
plotDnLocusBySampleSize(R=25, N=5000, r=1, q=0.0996, f=0.05, p_thres_denovo_locus=2.5e-06, name="Test3", col="#4daf4a") 

Run an example to estimate the power to detect a specific locus through an excess of rare protein truncating variants (PTVs) in autism spectrum disorder as the relative risk is varied:

  • Maximum relative risk (R) = 2.5
  • Sample size (N) = 20000 cases
  • Case to control ratio (r) = 1
  • Number of possible rare variants per gene (s) = 615
  • Number of possible rare PTVs per gene (f_gene) = 123
  • Proportion of rare PTVs that mediate risk (f) = 0.05
  • Prevalence of autism (K) = 0.01
  • Mean allele frequency of rare variants (AF_bar) = 0.001
  • Number of replicate simulations (N_rep) = 50
  • P-value threshold (p_thres_cc_locus_single) = 1.7e-11
  • Prefix for PDF of power (name) = Test4
  • Color of line in PDF (col) = "#4daf4a"
plotCcLocusSingleByRelativeRisk(R=2.5, N=20000, r=1, s=615, f_gene=123, f=0.05, K=0.01, AF_bar=0.001, N_rep=10, p_thres_cc_locus_single=1.7e-11, name="Test4", col="#4daf4a") 

For the analyses used to make the figures in the WGSPD manuscript, please see th vignettes.

About

Power calculations for whole genome sequencing

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • R 100.0%