Options

Options Attribute Default Detail
--single-trait flag - to implement single trait model.
--multi-trait flag - to implement multiple traits model.
--make-xrm flag - to construct and output relationship matrix.
--trans-xrm flag - to transform file format of relationship matrix.
--mme flag - to solve the mixed model equation with provided (co)variance components.
--snp-effect flag - to compute and output snp effect.
--maf flag - to compute the minor allele frequency.
--mating flag - to implement genomic mating.
--inbreeding flag - to calculate the inbreeding coefficient.
--pheno file - the path and name of phenotype file, header should be included, missing value should be represented by "NA".
--pheno-pos int 2 the position of the analyzed phenotype in columns of phenotype file. For multiple traits, please use comma as separator.
--pedigree file - the path and name of pedigree file, header should be included, the order of columns should be "id", "father", "mother".
--bfile file - the prefix of genotype files in plink binary format.
--gebv file - the estimated genomic breeding values, the first column should be id, header should be included.
--score file - the estimated SNP effects, the first column should be SNP id, the rest columns are the SNP effects, header should be included.
--xrm file - the prefix of binary relationship matrix (*.bin and *.id). For multiple relationship matrix, please use comma as separator.
--dcovar int - the position of the discrete covariates (fixed effects, e.g. sex, year, ...) in columns of phenotype file (e.g. "--dcovar 4,5"). For interaction, please use colon as separator (e.g. "--dcovar 4,5,4:5").
--qcovar int - the position of the quantitative covariates (e.g. height, weight, ...) in columns of phenotype file.
--rand int - the position of the random effects in columns of phenotype file. For interaction, please use colon as separator.
--rand-gxe int:file - a mixture of digital value and prefix of XRM, specify genetic-environment interaction (e.g. "--rand-gxe 2:demo.GA").
--add flag - additive genetic effect.
--add-inv flag - whether to compute the inverse of additive relationship matrix.
--dom flag - dominant genetic effect.
--dom-inv flag - whether to compute the inverse of dominant relationship matrix.
--vc-priors double - the prior values of variance components for all random effects, please use comma as separator within a trait, and use space as separator among traits.
--covc-priors double - the prior values of co-variance components among traits for all 'xrm', please use comma as separator among traits, and use space as separator among 'xrm'.
--snp-weight file - the path and name of a file containing weights for all SNPs when constructing GRM and calculating SNP effects, the fist column shoud be SNP id.
--vc-method text AI the algorithms (AI, EM, EMAI, HE, HI) for variance components estimation.
--tol double 1.00E-06 the tolerance for convergence of variance components estimation.
--ai-maxit int 20 the maximum number of iteration for AIREML.
--em-maxit int 20 the maximum number of iteration for EMREML.
--alpha double 0.05 weight of pedigree based relationship matrix when constructing H matrix for SSGBLUP model.
--ridge-value double 0 ridge regression value for solving uninvertible square matrix.
--triangle flag - input/output a triangle matrix (in format of 'i j value').
--xrm-txt file - the file path and name of relationship matrix in text format.
--xrm-id file - the file path and name of id for all individuals.
--keep file - the path and name of a file listing individuals to be used in analysis.
--remove file - the path and name of a file listing individuals to be removed in analysis.
--extract file - the path and name of a file listing markers to be used in analysis.
--exclude file - the path and name of a file listing markers to be removed in analysis.
--pcg flag - to solve the system of linear equations A*x = b for x using the Preconditioned Conjugate Gradients Method.
--pcg-num int 0 the maximum number of iteration for Preconditioned Conjugate Gradients Method.
--step int 10000 the number of markers that are loaded into RAM at a time when constructing GRM, smaller step helps to reduce the memory cost, but would cause lower computational efficiency.
--float-prec flag - whether to use float precision, it can help to reduce the memory cost in analysis.
--threads int 0 the number of threads used for analysis.
--out text hiblup the prefix of output files.