GWASTools (version 1.12.2)

ncdfImputedDosage: Create a NetCDF file with imputed dosages - deprecated

Description

Deprecated - use imputedDosageFile This function creates a NetCDF file and corresponding annotation for imputed dosages from IMPUTE2, BEAGLE, or MaCH.

Usage

ncdfImputedDosage(input.files, ncdf.filename, chromosome, input.type=c("IMPUTE2", "BEAGLE", "MaCH"),  input.dosage=FALSE, block.size=5000, snp.annot.filename="dosage.snp.RData", scan.annot.filename="dosage.scan.RData", verbose=TRUE)

Arguments

input.files

A character vector of input files. The first file should always be genotypes (either probabilities or dosages). Files for each input type should be as follows:

• IMPUTE2: 1) .gens, 2) .samples
• BEAGLE: 1) .grobs or .dose, 2) .markers
• MaCH: 1) .mlprob or .mldose, 2) .mlinfo, 3) file with columns named "SNP" and "position" giving base pair position of all SNPs

ncdf.filename

Character string with name of output NetCDF file.

chromosome

Chromosome corresponding to the SNPs in the genotype file. Character codes will be mapped to integer values as follows: "X"->23, "XY"->24, "Y"-> 25, "M","MT"->26.

input.type

Format of input files. Accepted file types are "IMPUTE2", "BEAGLE", and "MaCH".

input.dosage

Logical for whether the genotype file (input.files[1]) contains dosages. If FALSE (default), the genotype file is assumed to contain genotype probabilities.

block.size

Number of lines to read at once.

snp.annot.filename

Output .RData file for storing a SnpAnnotationDataFrame.

scan.annot.filename

Output .RData file for storing a ScanAnnotationDataFrame.

verbose

Logical for whether to print progress messages.

Details

Input files can contain either imputed dosages or genotype probabilities, specified by the input.dosage flag. In either case, the NetCDF file will store dosage of the A allele in the "genotype" variable. All SNPs are assumed to be on the same chromosome, which is indicated by the chromosome argument.

SNP and scan annotation are created from the input files and stored in RData format in snp.annot.filename and scan.annot.filename.

Currently supported input file types are IMPUTE2, BEAGLE, and MaCH.

References

IMPUTE2: http://mathgen.stats.ox.ac.uk/impute/impute_v2.html

BEAGLE: http://faculty.washington.edu/browning/beagle/beagle.html

MaCH: http://www.sph.umich.edu/csg/abecasis/MACH/tour/imputation.html

See Also

createDataFile, NcdfGenotypeReader, GenotypeData, assocTestRegression

Examples

## Not run: 
# ncfile <- tempfile()
# snpfile <- tempfile()
# scanfile <- tempfile()
# 
# # IMPUTE2
# probfile <- system.file("extdata", "imputation", "IMPUTE2", "example.chr22.study.gens",
#                         package="GWASdata")
# sampfile <- system.file("extdata", "imputation", "IMPUTE2", "example.study.samples",
#                         package="GWASdata")
# ncdfImputedDosage(input.files=c(probfile, sampfile), ncdf.filename=ncfile, chromosome=22,
#                   input.type="IMPUTE2", input.dosage=FALSE,
#                   snp.annot.filename=snpfile, scan.annot.filename=scanfile)
# nc <- NcdfGenotypeReader(ncfile)
# scanAnnot <- getobj(scanfile)
# snpAnnot <- getobj(snpfile)
# genoData <- GenotypeData(nc, scanAnnot=scanAnnot, snpAnnot=snpAnnot)
# geno <- getGenotype(genoData)
# getVariable(genoData, "alleleA")
# getVariable(genoData, "alleleB")
# 
# # association test with imputed dosages
# scanAnnot$status <- sample(0:1, nrow(scanAnnot), replace=TRUE)
# genoData <- GenotypeData(nc, scanAnnot=scanAnnot, snpAnnot=snpAnnot)
# assoc <- assocTestRegression(genoData, outcome="status", model.type="logistic",
#                              gene.action.list="additive", dosage=TRUE)
# head(assoc)
# close(genoData)
# 
# 
# # BEAGLE - genotype probabilities
# probfile <- system.file("extdata", "imputation", "BEAGLE", "example.hapmap.unphased.bgl.gprobs",
#                       package="GWASdata")
# markfile <- system.file("extdata", "imputation", "BEAGLE", "hapmap.markers",
#                     package="GWASdata")
# ncdfImputedDosage(input.files=c(probfile, markfile), ncdf.filename=ncfile, chromosome=22,
#                   input.type="BEAGLE", input.dosage=FALSE,
#                   snp.annot.filename=snpfile, scan.annot.filename=scanfile)
# 
# # BEAGLE - dosage
# dosefile <- system.file("extdata", "imputation", "BEAGLE", "example.hapmap.unphased.bgl.dose",
#                     package="GWASdata")
# ncdfImputedDosage(input.files=c(dosefile, markfile), ncdf.filename=ncfile, chromosome=22,
#                   input.type="BEAGLE", input.dosage=TRUE,
#                   snp.annot.filename=snpfile, scan.annot.filename=scanfile)
# 
# 
# # MaCH - genotype probabilities
# probfile <- system.file("extdata", "imputation", "MaCH", "mach1.out.mlprob",
#                         package="GWASdata")
# markfile <- system.file("extdata", "imputation", "MaCH", "mach1.out.mlinfo",
#                         package="GWASdata")
# posfile <- system.file("extdata", "imputation", "MaCH", "mach1.snp.position",
#                         package="GWASdata")
# ncdfImputedDosage(input.files=c(probfile, markfile, posfile), ncdf.filename=ncfile, chromosome=22,
#                   input.type="MaCH", input.dosage=FALSE,
#                   snp.annot.filename=snpfile, scan.annot.filename=scanfile)
# 
# # MaCH - dosage
# dosefile <- system.file("extdata", "imputation", "MaCH", "mach1.out.mldose",
#                         package="GWASdata")
# ncdfImputedDosage(input.files=c(dosefile, markfile, posfile), ncdf.filename=ncfile, chromosome=22,
#                   input.type="MaCH", input.dosage=TRUE,
#                   snp.annot.filename=snpfile, scan.annot.filename=scanfile)
# 
# unlink(c(ncfile, snpfile, scanfile))
# ## End(Not run)