qtl (version 1.01-9)

scanqtl: General QTL scan

Description

Performs a multiple QTL scan for specified chromosomes and positions or intervals, with the possible inclusion of QTL-QTL interactions and/or covariates.

Usage

scanqtl(cross, pheno.col=1, chr, pos, covar=NULL, formula,
        method=c("imp"), incl.markers=FALSE, verbose=TRUE)

Arguments

cross

An object of class cross. See read.cross for details.

pheno.col

Column number in the phenotype matrix which should be used as the phenotype.

chr

Vector indicating the chromosome for each QTL.

pos

List indicating the positions or intervals on the chromosome to be scanned. Each element should be either a single number (for a specific position) or a pair of numbers (for an interval).

covar

A data frame of covariates.

formula

An object of class formula indicating the model to be fitted. QTLs are indicated as Q1, Q2, etc. Covariates are indicated by their names in covar.

method

Indicates whether to use the EM algorithm or imputation. (Only imputation is implemented at this point.)

incl.markers

If FALSE, do calculations only at points on an evenly spaced grid.

verbose

If TRUE, give feedback about progress.

Value

• An object of class scanqtl. It is a multi-dimensional array of LOD scores, with the number of dimension equal to the number of QTLs specifed.

Details

The formula is used to specified the model to be fitted. In the formula, user has to use Q1, Q2, etc., or q1, q2, etc., to represent the QTLs, and the column names in the covariate data frame to represent the covariates. Only the interaction terms need to be specifed in the formula. The main effects of all input QTLs (as specified by chr and pos) and covariates (as specifed by covar) will be included by default. For example, if the formula is y~Q1*Q2*Sex, there are three elements in input chr and pos and Sex is one of the column names for input cov, the formula used in genome scan will be y ~ Q1 + Q2 + Q3 + Sex + Q1:Q2 + Q1:Sex + Q2:Sex + Q1:Q2:Sex.

The input pos is a list or vector to specify the position/range of the input chromosomes to be scanned. If it is a vector, it gives the specified position on the chromosomes. If it is a list, it will contain either the certain positions or a range on chromosomes. For example, the input chr is c(1, 6, 13). If pos is c(9.8, 34.0, 18.6), it means to scan chromosome 1 at 9.8cM, chromosome 6 at 34cM and chromosome 13 at 18.6cM. If pos is list(c(5,15), c(30,36), 18), it means to scan chromosome 1 from 5cM to 15cM, chromosome 6 from 30cM to 36cM, fix chromosome 13 at 18cM.

References

Sen, S. and Churchill, G. A. (2001) A statistical framework for quantitative trait mapping. Genetics 159, 371--387.

See Also

fitqtl, makeqtl

Examples

data(fake.f2)

# take out several QTLs
qc <- c(1, 8, 13)
fake.f2 <- subset(fake.f2, chr=qc)

# imputate genotypes
fake.f2 <- subset(fake.f2, ind=1:50)
fake.f2 <- sim.geno(fake.f2, n.draws=64, step=2, err=0.001)fake.f2 <- sim.geno(fake.f2, n.draws=8, step=5, err=0.001)

# 2-dimensional genome scan with additive 3-QTL model
pos <- list(c(15,35), c(45,65), 28)
result <- scanqtl(fake.f2, pheno.col=1, chr=qc, pos=pos,
                  formula=y~Q1+Q2+Q3)

# image of the results
# chr locations
chr1 <- as.numeric(matrix(unlist(strsplit(colnames(result),"@")),
                   ncol=2,byrow=TRUE)[,2])
chr8 <- as.numeric(matrix(unlist(strsplit(rownames(result),"@")),
                   ncol=2,byrow=TRUE)[,2])
# image plot
image(chr1, chr8, t(result), las=1, col=rev(rainbow(256,start=0,end=2/3)))

# do the same, allowing the QTLs on chr 1 and 13 to interact
result2 <- scanqtl(fake.f2, pheno.col=1, chr=qc, pos=pos,
                   formula=y~Q1+Q2+Q3+Q1:Q3)
# image plot
image(chr1, chr8, t(result2), las=1, col=rev(rainbow(256,start=0,end=2/3)))