The LD LASSO function

Description

This function implements a method for the automatic selection of parameters for the LD LASSO. It returns three solutions, the fused, cp-optimal, and unfused solutions. It also creates a matrix of solutions needed for creating the trace plot.

Usage

1
2
ld_lasso_method(block.obj, block.cood = NA, Xa = NA, Y = NA, bpmap = NA,
maxcol = 5e3, p.frac = 0.10, B = 5, s2low = 5e-3, s2high = 5e1, s2.vec.length = 4, null = FALSE)

Arguments

block.obj

An object of class gwaa.data from GenABEL.

block.cood

A vector of length p+1, where p is the number of SNPs. block.cood is an indicator vector that indicates block boundaries at all p+1 SNP bounded intervals. Use find.bounds to create this vector.

Xa

If block.obj is NA then a genotype matrix must be provided. Xa is a matrix of genotype values codes as 0, 1 or 2 for homozygous major, heterozygous, or homozygous minor, respectively.

Y

If block.obj is NA then a phenotype vector Y must be provided. Y is a vector of diagnoses, where 0 is non-diseased and 1 is diseased.

bpmap

A vector of map positions in terms of kB from the left boundary

maxcol

The upper limit on the number of columns in the constraint matrix. This is to prevent computational overload. Increasing maxcol may increase computation time and memory needed.

p.frac

The fraction of SNPs allowed in LASSO model under null hypothesis. This parameter is used in the function get.s1.

B

The number of bootstrap iterations for cp estimate.

s2low

The lower limit of the s2 vector

s2high

The upper limit of the s2 vector

s2.vec.length

The number of exponentially spaced values of s2.

null

A logical variable that indicates if analysis should be performed on permuted phenotype vector.

Details

Use function find.bounds to create block.cood vector with MATILDE MCMC methods. See ldlasso help page for details on the package MATILDE.

Value

beta1

The ld lasso solution with s2 that minimizes cp (cp-optimal solution)

beta2

The ld lasso solution for lower limit of s2 interval (fused solution)

beta3

The ld lasso solution for upper limit of s2 interval (unfused solution)

s2star

The s2 value that minimizes cp

cp.obj

A list that contains information used for cp estimation

log10p

A vector of log10 p values for test of allelic association

bpmap

A vector of map positions in base pairs

block.bounds.vec

A vector of block boundaries in kB from left boundary

s1

the LASSO constraint

B

Number of bootstrap samples

s2.vec.length

The length of the s2 vector

Author(s)

Samuel G. Younkin

References

Samuel Younkin, Joseph Nadeau, Robert Elston and J. Sunil Rao, "The Linkage Disequilibrium LASSO for SNP Selection in a Genetic Association Study of Late Onset Alzheimer Disease," Technical Report, 2010

See Also

ld_lasso

Examples

1
2
3
4
5
  # Load example data
  # data(ldlasso_example)
  
  # Run the method with low B and s2.vec.length first to test.
  # ldlasso.obj <- ld_lasso_method( block.obj, block.cood, B = 3, s2.vec.length = 2 )