RFGLS uses a generalized least-squares method to perform single-marker association analysis, in datasets of nuclear families containing parents, twins, and/or adoptees. It is designed for families of no greater than four members. When conducting association analysis with a large number of markers, as in GWAS, RFGLS uses rapid feasible generalized least-squares, an approximation to feasible generalized least-squares (FGLS) that considerably reduces computation time with minimal bias in p-values, and with negligible loss in power.
Fits a generalized least-squares regression model to test association between a quantitative phenotype and all SNPs in a genotype file, one at a time, via Rapid Feasible Generalized Least Squares. For each SNP, genotype is treated as a fixed effect, and the residual variance-covariance matrix is also estimated. In each trait-SNP association test, the fgls() function is used for parameter estimation.
This function creates the family-structure variables "FTYPE" (family-type) and "INDIV" (individual code) from available information in a pedigree file. Note that FSV.frompedi() is called internally by gls.batch() and gls.batch.get() when their argument input.mode is set to 3.
Jointly estimates the fixed-effects coefficients and residual variance-covariance matrix in a generalized least squares model by minimizing the (multivariate-normal) negative loglikelihood function, via optim() in the R base distribution. The residual variance-covariance matrix is block-diagonal sparse, constructed with bdsmatrix() from the bdsmatrix package.
providing 
.