stepwiseqtlShape: stepwiseqtlShape
In nnavarro/shapeQTL: shape QTL mapping experiment using R

Description Usage Arguments Details Value Author(s) References See Also Examples

View source: R/stepwiseqtl_shape.R

Runs a stepwise multiple QTL model search using penalized negative $log_10$ of the p-value

stepwiseqtlShape(cross, chr, pheno.col = 1, qtl, formula, max.qtl = 10,
  covar = NULL, refine.locations = TRUE, additive.only = FALSE,
  scan.pairs = FALSE, penalties, keeplodprofile = TRUE, keeptrace = FALSE,
  verbose = FALSE, test = "Pillai")

`cross`
`chr`
`pheno.col`
`qtl`	`summary.scanone` object or any data.frame with 'chr' 'pos' and 'lod' values. If missing a one-dimensional scan is run first.
`formula`
`max.qtl`
`covar`
`refine.locations`	Logical
`additive.only`	Logical. (Defauft FALSE). (Not as in `summary.scanone`). Here setting to TRUE may be used with f2 cross to force a diplotype model (pure additive)
`scan.pairs`	Logical. (Default FALSE) #keep it default. Not use so far
`penalties`	Genomewide threshold
`keeplodprofile`	Logical (Default TRUE)
`keeptrace`	Logical (Default FALSE)
`verbose`	Logical (Default FALSE)
`test`	Multivariate statistics (Default "Pillai". Others are : "Hotelling.Lawley", "Lik.ratio", "Goodall")

The function built multiple QTL models using penalized negative $log_10$ of the p-value (as a multivariate analogue of the penalized LOD score of the R/qtl package). The search is limited so far to a model without interaction

The function returns a qtl object.

Nicolas Navarro

Broman and Sen 2009. A guide to QTL mapping with R/qtl

stepwiseqtl

data(fake.bc)
fake.bc <- calc.genoprob(fake.bc, step=2.5)
fake.bc <- update.cross(fake.bc, phen2keep = names(fake.bc$pheno)) #Just update the class of fake.bc
covar <- fake.bc$pheno[, c('sex','age')]
out1 <- scanoneShape(fake.bc, pheno.col = 1:2, addcovar = covar,
test = "Pillai")
Q <- max(out1)
outStep <- stepwiseqtlShape(fake.bc, pheno.col = 1:2, qtl = Q, max.qtl = 3, covar = covar, penalties=2.5, verbose = TRUE)