lingweis: lingweis

In USCbiostats/LinGxEScanR: Run Linear Regression GWEIS Using VCF or Binary Dosage Files

Description

Run a linear gweis using genetic data from a binary dosage or vcf file

Usage

lingweis(
  data,
  ginfo,
  snps,
  gonly,
  ge,
  ggxe,
  gxe,
  joint,
  levene,
  testvalue,
  meta,
  outfile,
  outformat,
  minmaf,
  blksize
)

Arguments

data	Data frame containing the subject ID, phenotype and covariates. The first column must be a character value that contains the subject ID. The second column must be a numeric value and contain the outcome. The remaining columns must be numeric and contain the covariates to use in the analysis. The value in the last column is the covariate that will be used in the gene-environment interaction.
ginfo	Information about the binary dosage or vcf file returned from the BinaryDosage::getbdinfo or BinaryDosage::getvcfinfo routine
snps	The SNPs to be used in the scan. This may be an integer vector indicate which SNPs to use in the binary dosage file or a character vector of the SNP IDs to use. The value may also be "all", indicating to use all SNPs. The default value is "all".
gonly	The tests to perform on the beta_g parametetfor the model with the gene and all the covariates except the one used in the gene-environement interaction. The value must be a character array containing any or all of the following values, "fit", "lrt", "score", "robustscore", "Wald", and "robustWald". This value can also be "all" or "none".
ge	The tests to perform on the beta_g parameter from the model with the all the covariates and the gene. The value has the sames requirements as the gonly value.
ggxe	The test to perform on the beta_g parameter from the model that contains all the covariates, the gene, and the gene-environment interaction. The values has the same requirements as the gonly value.
gxe	The test to perform on the beta_gxe parameter from the model that contains all the covariates, the gene, and the gene-environment interaction. The values has the same requirements as the gonly value.
joint	The two degree of tests to perform on the beta_g and beta_gxe parameters from the model that contains all the covariates, the gene, and the gene-environment interaction. The values has the same requirements as the gonly value.
levene	Logical array indicating which Levene tests to run. The first value indicates to run the Levene test with the interaction covariate and the second value indicates to run the Levene test without the interaction covariate. This value may be a logical array of length one or two. If only one value is passed it is used for both tests.
testvalue	The value to output for the test. Can be "stat", "p", or "both". "stat" indicates to output the test statistic. "p" indicates to output the p-value. "both" indicates to output the test statistic and the p-value.
meta	Indicator to output the values needed for a meta-analysis.
outfile	The file name for the results Can be blank. If the value is "", the results are returned as a data frame. Default value is ""
outformat	Format of the output file. Can either be "text" or "RDS".
minmaf	Minimum minor allele frequency of SNPs to include in analysis. SNPS that have less than 20 minor alleles observed will be excluded from the analysis regardless of the value of minmaf. A value of 0 indicates to use all the SNPs that have 20 minor alleles observed. Default value is 0.
blksize	Size of blocks of SNPs to read in at one time. Larger blocks can improve overall speed but require larger amounts of computer memory. A value of 0 indicates to use the recommended block size. Default value is 0.

Value

0

Examples

bdinfo <- readRDS(system.file("extdata/pdata_4_1.bdinfo", package = "GxEScanR"))
covdata <- readRDS(system.file("extdata/covdata.rds", package = "GxEScanR"))

results <- gweis(data = covdata, ginfo = bdinfo)

USCbiostats/LinGxEScanR documentation built on Feb. 24, 2022, 12:43 p.m.