MDR (version 1.2)

mdr.hr: Function to estimate the accuracy of an MDR model given high-risk/low-risk status of genotype combinations

Description

Determines the balanced accuracy (mean of sensitivity and specificity) of an MDR model (specified combination of loci and high-risk/low-risk genotype combinations) which minimize balanced accuracy. Is used to determine prediction error estimates in cross-validation (mdr.cv).

Usage

mdr.hr(split, model, hr, genotype = c(0, 1, 2))

Arguments

split

the dataset; an n by (p+1) matrix where the first column is the binary response vector (coded 0 or 1) and the remaining columns are the p SNP genotypes (coded numerically)

model

a numeric vector of the final MDR model loci

hr

vector of binary indicators for high-risk/low-risk of the genotype combinations of the final model loci

genotype

a numeric vector of possible genotypes arising in split; default is c(0,1,2), but this vector can be longer or shorter depending on if more or fewer than three genotypes are possible

Value

List containing:

balanced accuracy

Balanced accuracy estimate (the mean of sensitivity and specificity) of the specified MDR model

...

Details

When determining the high-risk/low-risk status of a genotype combination, the order of combinations uses the convention that the genotypes of the first locus vary the most, based on the function expand.grid. For instance, with 3 genotypes (0,1,2), a two-way interaction results in the following 9 combinations: (0,0), (1,0), (2,0), (0,1), (1,1), (2,1), (0,2), (1,2), (2,2).

References

Ritchie et al (2001). Multifactor-dimensionality reduction reveals high-order interactions among estrogen-metabolism genes in sporadic breast cancer. Am J Hm Genet 69, 138-147.

Velez et al (2007). A balanced accuracy function for epistasis modeling in imbalanced datasets using multifactor dimensionality reduction. Genet Epidemiol 31, 306-315.

See Also

mdr,mdr.cv,mdr.3WS

Examples

#load test data
data(mdr1)

#split data into training and testing sets
train<-mdr1[1:125,]
test<-mdr1[-(1:125),]

#define matrix of all two-way combinations of 15 SNPs; this 105 by 2 matrix defines the 105 combinations of two-way interactions to consider 
loci<-t(combn(15,2)) 

#this runs mdr on the training data, considering the two-way combinations in 'loci', saving the top model, and defining the threshold as 1 since the data is balanced
fit<-mdr(train,loci,x=1,ratio=1) 

#estimate balanced accuracy given the MDR best model
acc<-mdr.hr(test,model=fit$models, hr=fit$high)

print(acc)