GWAS_scan: GWAS scan
In vincentgarin/GWASToolBox: GWAS analysis

Description Usage Arguments Details Value Author(s) References Examples

Perform a SNP genome wide association study using optional principal components terms or a kinship matrix to correct for the genetic background.

1
2
3

GWAS_scan(gp, trait = 1, model = "kinship", thre_cof = 6, win_cof,
  nPC = 2, K_i = TRUE, weights = NULL, power = -1, mk.sel = NULL,
  verbose = FALSE)

`gp`	`gpData` object with elements geno coded 0 1 2, map with marker position in bp or cM and phenotype.
`trait`	`Numerical` or `character` indicator to specify which trait of the gp object should be used. Default = 1.
`model`	`Character` argument specifying the type of model that is used to perform the GWAS analysis: 1) 'null' for a simple model with only the QTL position; 2) 'cofactors' for a model including QTL and cofactors; 3) 'PC' model with genetic background correction using the `nPC` first principal components; 4) 'kinship', model with kinship genetic background correction. Default = 'kinship'.
`thre_cof`	`Numeric` -log10(pval) above which a cofactor is selected. Default = 6.
`win_cof`	`Numeric` value indicating the minimum distance between two cofactors positions.
`nPC`	Optional number of principal components for genetic background correction if the 'PC' model is selected. Default = 2.
`K_i`	`Logical` specifying if the kinship correction should be done by removing the markers of the scanned chromosome. Default = TRUE.
`weights`	(Not available for the moment) object of class `LD_wgh` obtained by the function LDAK_weights() representing a data.frame with two columns: the marker identifier and the LD adjusted weights. These weight will be used to compute a LDAK as defined by Speed et al. (2012) for the genetic background adjustement. Default = NULL.
`power`	`Numerical` value specifying the value of the parameter for marker scores standardization. The column of the marker matrix (X.j) are multiplied by var(X.j)^(power/2). It correspond to alpha in the formula. Default = -1.
`mk.sel`	`Character vector` specifying a list of marker to use for the kinship matrix computation. By default, the function use all markers of the `gp`.
`verbose`	`Logical` indicating if function outputs should be printed. Default = FALSE.

The function is a wrapper for the GWAS function from the sommer package (Covarrubias-Pazaran, 2016). The model is fitted using the EMMA algorithm proposed by Kang et al. (2008).

By default, the kinship matrix is computed using the method of Astle and Balding (2009). It is possible to compute a linkage disequilibrium adjusted kinship (LDAK) kinship matrix using the method of Speed et al. (2012) by introducing the weights computed with the function LDAK_weights() (Not available for the moment). The model can be fitted using the kinship containing all markers or removing the markers of the scanned chromosome (K_i = TRUE).

Return:

G_res

Object of class G_res representing a data.frame with four columns: marker identifier, chromosome, position in cM or bp and -log10(p-value).

Vincent Garin

Astle, W., & Balding, D. J. (2009). Population structure and cryptic relatedness in genetic association studies. Statistical Science, 451-471.

Covarrubias-Pazaran G. 2016. Genome assisted prediction of quantitative traits using the R package sommer. PLoSONE 11(6):1-15.

Kang, H. M., Zaitlen, N. A., Wade, C. M., Kirby, A., Heckerman, D., Daly, M. J., & Eskin, E. (2008). Efficient control of population structure in model organism association mapping. Genetics, 178(3), 1709-1723.

Speed, D., Hemani, G., Johnson, M. R., & Balding, D. J. (2012). Improved heritability estimation from genome-wide SNPs. The American Journal of Human Genetics, 91(6), 1011-1021.