trio.prepare: Generate Trio Data Format Suitable for Trio Logic Regression

In trio: Testing of SNPs and SNP Interactions in Case-Parent Trio Studies

Description

This function transforms case-parent data into a format suitable as input for trio logic regression. The function can also be used for the imputation of missing genotypes in case-parent data, while taking the existing SNP block structure into account.

Usage

trio.prepare(trio.dat, freq=NULL, blocks=NULL, logic=TRUE, ...)

Arguments

trio.dat	An object returned from the function trio.check.
freq	An optional data frame specifying haplotype blocks and frequencies. For an example, see the data frame simuBkMap contained in this package. If provided, the following argument blocks will be ignored. The object must have three columns in the following order: block identifiers (key), haplotypes (hap), and haplotype frequencies (freq). The block identifiers must be unique for each block. For each block, the haplotypes must be encoded as a string of the integers 1 and 2, where 1 refers to the major allele and 2 refers to the minor allele. The respective haplotype frequencies will be normalized to sum one.
blocks	An optional vector of integers, specifying (in sequence) the lengths of the linkage disequilibrium blocks. The sum of these integers must be equal to the total numbers of SNPs in the data set used as input. Using the integer 1 for SNPs not contained in LD blocks is required if this argument is used. If both arguments freq and blocks are NULL, complete linkage equilibrium is assumed (i.e., no correlation between the genotypes).
logic	A logical value indicating whether the trio data are returned with genotypes in dominant and recessive coding, suitable as input for trio logic regression (TRUE), or if the imputed data should be returned in genotype format, using one variable per SNP (FALSE).
...	Optional arguments that can be passed to function haplo.em.

Details

To create the genotypes for the pseudo-controls it is necessary to take the LD structure of the SNPs into account. This requires information on the LD blocks. It is assumed that the user has already delineated the block structure according to his or her method of choice. The function trio.prepare, which operates on an output object of trio.check, accepts the block length information as an argument. If this argument is not specified, a uniform block length of 1 (i.e., no LD structure) is assumed. If the haplotype frequencies are not specified, they are estimated from the parents' genotypes using the function haplo.em. The function then returns a list that contains the genotype information in binary format, suitable as input for trio logic regression. Since trio logic regression requires complete data, the function trio.prepare also performs an imputation of the missing genotypes. The imputation is based on the estimated or supplied haplotype information.

Value

bin	A matrix suitable as input for trio logic regression. The first column specifies the cases and pseudo-controls as required by logic regression using conditional logistic regression (the integer 3 for the probands followed by three zeros indicating the pseudo-controls). The following columns specify the (possibly imputed) genotypes in dominant and recessive coding, with two binary variables for each SNP. This is returned only if logic = TRUE.
trio	A data frame with imputed SNPs in genotype format derived from the input. This is returned only if logic = FALSE.
miss	A data frame with five columns indicating the missing genotypes in the input object. The five columns of the data frame refer to the family id (famid), the individual id (pid), the genotype (snp), the row numbers (r), and the column numbers (c). This element will be NULL if there are no missing data.
freq	The estimated or supplied haplotype information, in the same format as described in the Arguments above.

Acknowledgments

Support was provided by NIH grants R01 DK061662 and HL090577.

Author(s)

Qing Li, [e-mail]

References

Li, Q., Fallin, M.D., Louis, T.A., Lasseter, V.K., McGrath, J.A., Avramopoulos, D., Wolyniec, P.S., Valle, D., Liang, K.Y., Pulver, A.E., and Ruczinski, I. (2010). Detection of SNP-SNP Interactions in Trios of Parents with Schizophrenic Children. Genetic Epidemiology, 34, 396-406.

See Also

trio.check, haplo.em

Examples

data(trio.data)
trio.tmp <- trio.check(dat=trio.ped1)
trio.bin <- trio.prepare(trio.dat=trio.tmp, blocks=c(1,4,2,3))
trio.bin$bin[1:8,]

trio documentation built on Nov. 8, 2020, 7:41 p.m.