GWASTools: Tools for Genome Wide Association Studies

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

These functions create a simulated genotype or intensity file for test and examples.

1
2
3

simulateGenotypeMatrix(n.snps=10, n.chromosomes=10,
                       n.samples=1000, filename,
                       file.type=c("gds", "ncdf"), silent=TRUE)

`n.snps`	An integer corresponding to the number of SNPs per chromosome, the default value is 10. For this function, the number of SNPs is assumed to be the same for every chromosome.
`n.chromosomes`	An integer value describing the total number of chromosomes with default value 10.
`n.samples`	An integer representing the number of samples for our data. The default value is 1000 samples.
`filename`	A string that will be used as the name of the file. This is to be used later when opening and retrieving data generated from this function.
`file.type`	The type of file to create ("gds" or "ncdf")
`silent`	Logical value. If `FALSE`, the function returns a table of genotype counts generated. The default is `TRUE`; no data will be returned in this case.

The resulting netCDF file will have the following characteristics:

Dimensions:

'snp': n.snps*n.chromosomes length

'sample': n.samples length

Variables:

'sampleID': sample dimension, values 1-n.samples

'position': snp dimension, values [1,2,...,n.chromosomes] n.snps times

'chromosome': snp dimension, values [1,1,...]n.snps times, [2,2,...]n.snps times, ..., [n.chromosomes,n.chromosomes,...]n.snps times

'genotype': 2-dimensional snp x sample, values 0, 1, 2 chosen from allele frequencies that were generated from a uniform distribution on (0,1). The missing rate is 0.05 (constant across all SNPs) and is denoted by -1.

'quality': 2-dimensional snp x sample, values between 0 and 1 chosen randomly from a uniform distribution. There is one quality value per snp, so this value is constant across all samples.

'X': 2-dimensional snp x sample, value of X intensity taken from a normal distribution. The mean of the distribution for each SNP is based upon the sample genotype. Mean is 0,2 if sample is homozygous, 1 if heterozygous.

'Y': 2-dimensional snp x sample, value of Y intensity also chosen from a normal distribution, where the mean is chosen according to the mean of X so that sum of means = 2.

simulateGenotypeMatrix returns a table of genotype calls if the silent variable is set to FALSE, where 2 indicates an AA genotype, 1 is AB, 0 is BB and -1 corresponds to a missing genotype call.

simulateIntensityMatrix returns a list if the silent variable is set to FALSE, which includes:

`het`	Heterozygosity table
`nmiss`	Number of missing values

A file is created and written to disk.

Caitlin McHugh

GdsGenotypeReader, GdsIntensityReader, NcdfGenotypeReader, NcdfIntensityReader

filenm <- tempfile()

simulateGenotypeMatrix(filename=filenm )

file <- GdsGenotypeReader(filenm)
file #notice the dimensions and variables listed

genot <- getGenotype(file)
table(genot) #can see the number of missing calls

chrom <- getChromosome(file)
unique(chrom) #there are indeed 10 chromosomes, as specified in the function call

close(file)

simulateIntensityMatrix(filename=filenm, silent=FALSE )

file <- GdsIntensityReader(filenm)
file #notice the dimensions and variables listed

xint <- getX(file)
yint <- getY(file)
print("Number missing is: "); sum(is.na(xint))

chrom <- getChromosome(file)
unique(chrom) #there are indeed 10 chromosomes, as specified in the function call

close(file)

unlink(filenm)