calcCis: Calculate all confidence intervals from a QTL model
In jtlovell/qtlTools: Data Processing and Plotting in Association with R/qtl

Description Usage Arguments Details Value Examples

calcCis Takes a multiple QTL model or scanone and permutations output and returns a dataframe of confidence intervals.

1
2
3

calcCis(cross, mod = NULL, s1.output = NULL, pheno.col = NA,
  perm.output = NULL, qtlnames = NULL, lodint = TRUE, drop = 1.5,
  prob = 0.95, expandtomarkers = FALSE, chr = NULL, ...)

`cross`	The QTL cross object.
`mod`	A QTL model containing LOD profiles. Can be generated by stepwiseqtl (with keepLodProfile = TRUE), or by refineqtl.
`s1.output`	Output from a scanone call. Not needed if qtl models are supplied to mod. If mod is not supplied, it is necessary to provide both s1.output and perm.output.
`pheno.col`	If passing a QTL model, specify the name of the phenotype.
`perm.output`	Output from a scanone call with n.perm > 1. Needs to have the identical parameters as s1.output.
`qtlnames`	An optional vector of user-specified QTL identifiers. Used only if mod is supplied.
`lodint`	Should lodint (default) or bayesint be used to estimate confidence intervals?
`drop`	If lodint = TRUE, what drop threshold should be used?
`prob`	If lodint = FALSE, what significance should be used for bayesint?
`expandtomarkers`	Logical. Should the intervals be expanded to the nearest real marker?
`...`	Additional arguments passed on to pullSigQTL, such as the alpha threshold for a qtl.

This function iterates through the qtl in the model, returning confidence intervals.

A dataframe of confidence intervals, including the maximum LOD score, bounding markers and positions.

library(qtlTools)
data(fake.bc)
cross<-fake.bc
cross <- calc.genoprob(cross, step=2.5)

# Make a QTL object and formula
mod <- makeqtl(cross, chr = c(2,5), pos = c(40,25), what = "prob")
sex <- data.frame(sex = cross$phe$sex)
nform <- "y ~ Q1 + Q2 + Q1*sex + sex"

#Calculate lodprofiles for confidence interval estimation
mod <- refineqtl(cross, mod, pheno.col = "pheno1",
  qtl = mod, formula = nform, covar = sex, method="hk")
calcCis(cross = cross, mod=mod)
## Not run: 
s1<-scanone(cross, method="hk", pheno.col=c("pheno1", "pheno2"))
perm<-scanone(cross, n.perm=100, method="hk",pheno.col=c("pheno1", "pheno2"), verbose=FALSE)
calcCis(cross,s1.output=s1, perm.output=perm)

## End(Not run)