[Feature] make it easier to simulate multiple famlies at once

[smasongarrison]

make it easier to simulate multiple famlies at once, right now the solution is # True variance components
true_var <- list(
ad2 = 0.70, cn2 = 0.00, ee2 = 0.20,
ce2 = 0, mt2 = 0.10, dd2 = 0, am2 = 0
)

n_families <- 15
start_vars <- list(
ad2 = 0.3, # additive genetic
cn2 = 0.1, # common nuclear environment
ce2 = 0, # common extended (not estimated here)
mt2 = 0.1, # mitochondrial
dd2 = 0, # dominance (not estimated here)
am2 = 0, # A x Mt interaction (not estimated here)
ee2 = 0.5 # unique environment
)

Pre-allocate storage

add_list <- vector("list", n_families)
cn_list <- vector("list", n_families)
mt_list <- vector("list", n_families)
obs_ids_list <- vector("list", n_families)
pheno_list <- vector("list", n_families)
ped_list <- vector("list", n_families)
for (i in seq_len(n_families)) {
ped_i <- simulatePedigree(kpc = 3, Ngen = 6, marR = 0.6)

A_i <- as.matrix(ped2add(ped_i, sparse = FALSE))
Cn_i <- as.matrix(ped2cn(ped_i, sparse = FALSE))
Mt_i <- as.matrix(ped2mit(ped_i, sparse = FALSE))
n_i <- nrow(A_i)

Implied covariance from true parameters

V_i <- true_var$ad2 * A_i +
true_var$cn2 * Cn_i +
true_var$mt2 * Mt_i +
true_var$ee2 * diag(1, n_i)

y_i <- rmvnorm(1, sigma = V_i)
y_binary_i <- ifelse(y_i > 0, 1, 0)

ped_i$y <- y_i
ped_i$y_binary <- y_binary_i

ids_i <- make.names(rownames(A_i))
rownames(A_i) <- colnames(A_i) <- ids_i
rownames(Cn_i) <- colnames(Cn_i) <- ids_i
rownames(Mt_i) <- colnames(Mt_i) <- ids_i

add_list[[i]] <- A_i
cn_list[[i]] <- Cn_i
mt_list[[i]] <- Mt_i
obs_ids_list[[i]] <- ids_i
pheno_list[[i]] <- matrix(y_binary_i, nrow = 1, dimnames = list(NULL, ids_i))
ped_list[[i]] <- ped_i

}