multic: Create a multic object

In multic: Quantitative Linkage Analysis Tools using the Variance Components Approach

Description

Calculate the polygenic and major gene models for quantitative trait linkage analysis using variance components approach.

Usage

multic(formula, data = sys.parent(), famid, id, dadid, momid, sex,
       mloci.out = NULL, share.out = "kinship", longitudinal = FALSE,
       subset = NULL, ascertainment = NULL,
       control = multic.control(...), ...)

Arguments

formula	a formula object, with the traits on the left of a ~ (tilde) operator and the covariates, separated by + operators, on the right. The traits may be a single numeric vector or a matrix. Commonly, traits are aggragated together using the cbind command. See the Examples section for examples.
famid	integer, numeric, or character vector specifying each individual's family identifier. Members of the same family must have the same famid and each family must have a unique famid. Any missing data will result in an error message and the termination of multic.
id	integer, numeric, or character vector specifying each individual's identifier. Members of the same family must have a unique id within the family. id does not have to be universally unique among all individuals. Any missing data will result in an error message and the termination of multic.
dadid	integer, numeric, or character vector specifying each individual's father identifier. This father identifier must have the same famid as the individual. Any missing data will result in an error message and the termination of multic.
momid	integer, numeric, or character vector specifying each individual's mother identifier. This mother identifier must have the same famid as the individual. Any missing data will result in an error message and the termination of multic.
sex	integer, numeric, or character vector specifying each individual's sex. Acceptable forms of sex-coding are "M", "m", or 1 for male and "F", "f", or 2 for female. Any missing data will result in an error message and the termination of multic.
data	data.frame in which to interpret the variables named in formula, famid, id, dadid, momid, sex, subset, and ascertainment arguments. If data is missing, the variables in formula should be in the search path.
mloci.out	a character value specifying a path to an mloci.out (or similarly formatted) file. Specifying a non-empty mloci.out file will allow multic to calculate sporadic models using the ibd (identity by decent) information in the mloci.out file. Due to the general size of mloci.out, it is often stored in .gz format. multic will manage this for the user. Whether the user specifies an mloci.out file with a .gz suffix or not will not effect how multic operates on the file. See solar2mloci for more details.
share.out	a character value specifying a path to a share.out (or similarly formatted) file. This file contains the amount of genetic material shared between each family member pair based on family structure only. It also contains boolean values to indicate whether two familiy members have a sibling-sibling, parent-parent, or parent-offspring relationships. Due to the general size of share.out, it is often stored in .gz format. multic will manage this for the user. Whether the user specifies the file wiht a .gz suffix format or not will not effect how multic operates on the file. See phi2share for more details.
longitudinal	logical flag: if TRUE, then fomula will be interpreted as a longitudinal model. In this case, the fomula argument requires special formatting as described. The number of traits on the left side of the ~ (tilde) is the number of time-points for multic to analyze. The number of covariates on the right side of the ~ (tilde) must be a multiple of the number of traits on the left side. That multiple is the amount of covariates to analyze at each time-point. All covariates for the first time point must be specified before any of the second, all second before any third, etc. See the Examples section for examples.
subset	a logical vector specifying which subset of the rows in data to use in the fit.
ascertainment	vector specifying each individual's ascertainment (effected) status. Acceptable forms of ascertainment are T, TRUE, or 1 for a proband (effected) and F, FALSE, or 0 for a non-proband (non-effected person).
control	list of iteration and algorithmic constants. See multic.control for their names and default values. These can also be given directly as arguments to multic itself, instead of through a multic.control object. If control is specified, the arguments specified in ... will not be used.
...	further arguments passed to multic.control to alter multic's default behavior.

Details

See the technical report.

Value

an object of class "multic". See multic.object for more details.

Side Effects

Many temporary files are created during multic's execution. These files are deleted afterwards (by default). If they are not deleted (due to a crash or some other unexpected action), use the included function clean() to delete them. Also, multic copies, gunzip's, and removes the copies of share.out and mloci.out (if specified).

References

Amos, C. I. (1994). "Robust variance-components approach for assessing genetic linkage in pedigrees." American Journal of Human Genetics 54(3): 535-543.

Almasy, L. and J. Blangero (1998). "Multipoint quantitative-trait linkage analysis in general pedigrees." American Journal of Human Genetics 62(5): 1198-1211.

See Also

multic.object, multic.control, phi2share, solar2mloci, solar2multic, sw2mloci

Examples

## Not run: 
# Call multic with a univariate formula with two covariates and no
# markers (no mloci.out argument).
fit.ibd.uni <- multic(k.trig ~ sex.x + agexam,
                      data = ped.phen.data,
                      famid, id, fa, mo, sex.x,
                      share.out = 'multicInput/share.out')

# Call multic with a bivariate formula with three covariates, no
# markers (no mloci.out argument), and calculate the family log
# likelihoods.
fit.ibd.bi <- multic(cbind(k.trig, k.chol) ~ sex.x + agexam + agexam2,
                     data = ped.phen.data,
                     famid, id, fa, mo, sex.x,
                     share.out = 'multicInput/share.out',
                     calc.fam.log.liks = TRUE)

# Call multic with a longitudinal formula with six covariates letting
# the kinship library calculate the share.out argument.
long.fit <- multic(cbind(sbpA, sbpB, sbpC) ~
                     sexA + ageA + bmiA + generA + ageAg + smkA +
                     sexB + ageB + bmiB + generB + ageBg + smkB +
                     sexC + ageC + bmiC + generC + ageCg + smkC,
                   data = long.data,
                   famid, id, dadid, momid, sex,
                   longitudinal = TRUE)

## End(Not run)

multic documentation built on Jan. 15, 2017, 9:30 a.m.