SimRVPedigree (version 0.4.4)

trim_ped: Trim pedigree based on proband recall

Description

Primarily intended as an internal function, trim_ped chooses a proband and trims relatives based on the proband's probability of recalling his or her relatives.

Usage

trim_ped(ped_file, recall_probs = NULL)

Arguments

ped_file

An object of class ped. A pedigree generated by sim_ped or sim_RVped, or an object created by the function new.ped. See details.

recall_probs

Numeric. The proband's recall probabilities for relatives, see details. If not supplied, the default value of four times kinship coefficient between the proband and the relative is used.

Value

ped_trim The trimmed pedigree.

Details

By default recall_probs is four times the kinship coefficient, as defined by Thompson (see references), between the proband and the probands relative, which results in a recall probability of \(2^{-(n-1)}\) for a relative of degree \(n\). Alternatively, the user may specify a list of recall probabilities of length \(l > 0\), in which case the first l-1 items in recall_probs are the respective proband recall probabilities for relatives of degree 1, 2, ..., l-1, and the lth item in recall_probs is the proband's recall probability for all relatives of degree l or greater. For example if recall_probs = c(1) all relatives will be recalled by the proband with probability 1.

Occasionally, a trimmed family member must be retained to ensure that the pedigree can be plotted. When this occurs, family members who share a non-zero kinship coefficient with the proband are censored of all pertinent information, and will always have the following qualities:

1. availability status = 0

2. affected status = NA

3. birth year = NA

4. onset year = NA

5. death year = NA

6. RR = NA

Users who wish to use trim_ped for pedigrees not generated by sim_ped or sim_RVped must use new.ped to create an object of class ped. The ped object must contain the following variables for each pedigree member:

name	type	description
FamID	numeric	family identification number
ID	numeric	individual identification number
dadID	numeric	identification number of father
momID	numeric	identification number of mother
sex	numeric	gender identification; if male sex = 0, if female sex = 1
affected	logical	disease-affection status:
proband	logical	a proband identifier: proband = TRUE if the individual is the proband, and FALSE otherwise.
		affected = TRUE if affected by disease , and FALSE otherwise,
birthYr	numeric	the individual's birth year.
onsetYr	numeric	the individual's disease onset year, when applicable.
deathYr	numeric	the individual's death year, when applicable.
RR	numeric	the individual's relative risk of disease.
available	logical	availibility status;
		available = TRUE if available, and FALSE otherwise.

References

Nieuwoudt, Christina and Jones, Samantha J and Brooks-Wilson, Angela and Graham, Jinko (2018). Simulating Pedigrees Ascertained for Multiple Disease-Affected Relatives. Source Code for Biology and Medicine, 13:2.

Thompson, E. (2000). Statistical Inference from Genetic Data on Pedigrees. NSF-CBMS Regional Conference Series in Probability and Statistics, 6, I-169.

See Also

sim_RVped, sim_ped, new.ped

Examples

# NOT RUN {
#Read in example pedigree to trim
data(EgPeds)
egPeds <- new.ped(EgPeds)

#plot example_ped using kinship2
plot(subset(egPeds, FamID == 1), location = "topright", cex = 0.85)
mtext("Original Pedigree", side = 3, line = 2)


## Trim pedigree examples
# Illustrate the effect of various settings for recall_probs
Recall_Probabilities <- list(c(1),
                             c(1, 0.5),
                             c(1, 0.25, 0.1))


for (k in 1:length(Recall_Probabilities)) {
   set.seed(2)
   #trim pedigree
   TrimPed <- trim_ped(ped_file = subset(egPeds, FamID == 1),
                       recall_probs = Recall_Probabilities[[k]])

   plot(TrimPed, location = "topright", cex = 0.85)
   mtext(paste0("recall_probs = (", sep = "",
                paste(Recall_Probabilities[[k]], collapse = ", "), ')' ),
                side = 3, line = 2 )
}


# }