cpgen (version 0.1)

cgrm: Genomic Relationship Matrices

Description

Based on a coefficient-matrix (i.e. marker matrix) $\mathbf{X}$ that will be scaled column-wise, a weight-vector $\mathbf{w}$ and a shrinkage parameter $\lambda$, cgrm returns the following similarity matrix: $$\mathbf{G} = (1-\lambda) \frac{\mathbf{X D X}^{'}}{\sum\mathbf{w}} + \mathbf{I}\lambda$$

where $\mathbf{D}$ = $diag(\mathbf{w})$. A weighted genomic relationship matrix allows running TA-BLUP as described in Zhang et al. (2010).

Usage

cgrm(X, w = NULL, lambda=0)

Arguments

X

coefficient matrix

w

numeric vector of weights for every column in X

lambda

numeric scalar, shrinkage parameter

Value

• Similarity matrix with dimension nrow(X)

Details

...

References

de los Campos, G., Vazquez, A.I., Fernando, R., Klimentidis, Y.C., Sorensen, D., 2013. "Prediction of Complex Human Traits Using the Genomic Best Linear Unbiased Predictor". PLoS Genetics 9, e1003608. doi:10.1371/journal.pgen.1003608

Zhang Z, Liu J, Ding X, Bijma P, de Koning D-J, et al. (2010) "Best Linear Unbiased Prediction of Genomic Breeding Values Using a Trait-Specific Marker-Derived Relationship Matrix". PLoS ONE 5(9): e12648. doi:10.1371/journal.pone.0012648

See Also

cgrm.A, cgrm.D.

Examples

# generate random data
rand_data(500,5000)

weights <- (cor(M,y)**2)[,1]

G <- cgrm(M,weights,lambda=0.01)