Skip to contents

Create an mxModel for a pedigree

Source: R/buildmxPedigrees.R
buildPedigreeModelCovariance.Rd

This function builds an OpenMx model for a pedigree with specified variance components. It requires the OpenMx package.

Usage

buildPedigreeModelCovariance(
  vars = list(ad2 = 0.5, dd2 = 0.3, cn2 = 0.2, ce2 = 0.4, mt2 = 0.1, am2 = 0.25, ee2 =
    0.6),
  Vad = TRUE,
  Vdd = FALSE,
  Vcn = TRUE,
  Vce = TRUE,
  Vmt = TRUE,
  Vam = FALSE,
  Ver = TRUE,
  temporal = FALSE,
  p_hist = 0,
  components = c("a", "e"),
  start_beta0 = 0.5,
  start_beta_time = 0,
  start_gamma = 0,
  time_point_max = NULL,
  mean_degree = 0,
  start_mean = 0,
  mean_hist_free = FALSE,
  start_mean_hist = 0,
  lbound = 1e-10
)

Arguments

vars

A named list or vector of initial variance component values. Names should include ad2 (additive), dd2 (dominance), cn2 (common nuclear), ce2 (common extended), mt2 (mitochondrial), am2 (additive-mitochondrial interaction), and ee2 (unique environment). Default values are provided. Only used when temporal = FALSE.

Vad

Logical. Include additive genetic variance component. Default is TRUE.

Vdd

Logical. Include dominance genetic variance component. Default is FALSE.

Vcn

Logical. Include common nuclear family environment variance component. Default is TRUE.

Vce

Logical. Include common extended family environment variance component. Default is TRUE.

Vmt

Logical. Include mitochondrial genetic variance component. Default is TRUE.

Vam

Logical. Include additive by mitochondrial interaction variance component. Default is FALSE.

Ver

Logical. Include unique environmental variance component. Default is TRUE.

temporal

Logical. If TRUE, build a time-varying covariance sub-model in which each component's weight is a function of birth year (and, optionally, historical moderators) rather than a fixed scalar. See p_hist, components, start_beta0, start_beta_time, start_gamma, and time_point_max. Default is FALSE.

p_hist

Integer. Number of historical moderator columns. Only used when temporal = TRUE. Default is 0.

components

Character vector of component keys to include when temporal = TRUE (any of "a", "d", "cn", "ce", "mt", "am", "e"). Default is c("a", "e").

start_beta0

Numeric. Starting value for each component's intercept loading. Only used when temporal = TRUE. Default is 0.5.

start_beta_time

Numeric. Starting value for each component's time-slope loadings. Only used when temporal = TRUE. Default is 0.

start_gamma

Numeric. Starting value for historical-moderator loadings. Only used when temporal = TRUE. Default is 0.

time_point_max

Integer. Degree of the polynomial birth-year basis. Only used when temporal = TRUE. Default is 3.

mean_degree

Integer. Degree of the polynomial temporal mean. The default, 0, estimates a constant mean. Must not exceed time_point_max.

start_mean

Numeric. Starting value for the mean's intercept. Set this near the observed phenotype mean.

mean_hist_free

Logical, length 1 or p_hist. Whether historical moderators can shift the phenotypic mean. Defaults to FALSE, fixing those shifts at zero.

start_mean_hist

Numeric. Starting value for the free elements of G_mean.

lbound

Numeric. A lower bound for the variance components to ensure they remain positive during optimization. Default is 1e-10

Value

An OpenMx model representing the pedigree with specified variance components.