cshare {SHARE}R Documentation

Stepwise search for the most informative haplotypes

Description

The cshare function seeks the most informative set of SNPs for genetic association in a targeted region by growing/shrinking haplotypes with one more/less SNP in a stepwise fashion, and comparing prediction errors of different models via cross-validation or BIC.

Usage

cshare(haploObj, status, nfold = 10, maxsnps, tol = 1e-08,
       verbose = FALSE, ModSelMethod = c("Cross-Val", "BIC"),
       Minherit = c("additive", "dominant", "recessive"))

Arguments

haploObj A object of calss "haplo". haploObj could be generated by the function haplo.
status A character string indicating the column name of the phenotype in haploObj@pheno to be used as the clinical status in the analysis.
nfold An integer that determines how many folds in the cross-validation, if ModSelMethod="Cross-Val"
maxsnps An integer that determines the maximal number of SNPs to be chosen. The default is 6.
tol The convergence parameters in haplotype logistic regression
verbose TRUE/FALSE to decide whether to create log file for debug
ModSelMethod Model selection method. Possible methods are "Cross-Val" for cross-validation, and "BIC" for Bayesian information criterion
Minherit Mode of inheritance. Possible mode are "additive", "dominant", and "recessive"

Details

This function takes input from a phased genotype dataset and case-control status, performs stepwise search for the most informative haplotypes based on deviance criterion, either by cross-validation or by BIC. The function output the prediction deviances of a ladder of models up to the size of "maxsnps" values and best haplotype model selected.

Value

cshare returns a object of class "share"

Author(s)

James Y. Dai

References

Dai, J. Y., LeBlanc, M., Smith, N. L., Psaty, B. M. and Kooperberg, C. (2009). SHARE: an adaptive algorithm to select the most informative set of SNPs for genetic associat ion. Biostatistics, In Press.

See Also

haplo, share-class

Examples

## See vignette for more details
## Not run: 
unphasedKerem <- list()
unphasedKerem[["Cross-Val"]] <- cshare(unphasedHaplo, status="CF",
                                       nfold=20, maxsnps=5,
                                       ModSelMethod="Cross-Val",
                                       Minherit="additive")
unphasedKerem[["Cross-Val"]]

unphasedKerem[["BIC"]] <- cshare(unphasedHaplo, , status="CF",
                         maxsnps=5, ModSelMethod="BIC", Minherit="additive", verbose=1)
unphasedKerem[["BIC"]]
## End(Not run)

[Package SHARE version 1.0.4 Index]