LTFGRS: Implementation of Several Phenotype-Based Family Genetic Risk Scores

Description

Implementation of several phenotype-based family genetic risk scores with unified input data and data preparation functions to help facilitate the required data preparation and management. The implemented family genetic risk scores are the extended liability threshold model conditional on family history (LT-FH++) from Pedersen (2022) doi:10.1016/j.ajhg.2022.01.009 and Pedersen (2023) https://www.nature.com/articles/s41467-023-41210-z, Pearson-Aitken Family Genetic Risk Scores (PA-FGRS) from Krebs (2024) doi:10.1016/j.ajhg.2024.09.009, and family genetic risk score by Kendler (2021) doi:10.1001/jamapsychiatry.2021.0336.

Author(s)

Maintainer: Emil Michael Pedersen emp@ph.au.dk

Authors:

• Jette Steinbach jst@econ.au.dk

• Morten Dybdahl Krebs morten.dybdahl.krebs@regionh.dk

Other contributors:

• Lucas Rasmussen lar.ncrr@au.dk [contributor]

See Also

Useful links:

• https://emilmip.github.io/LTFGRS/

• Report bugs at https://github.com/EmilMiP/LTFGRS/issues

Wrapper around the Gibbs Sampler that returns formatted liability estimates for the proband

Description

Wrapper around the Gibbs Sampler that returns formatted liability estimates for the proband

Usage

Gibbs_estimator(cov, tbl, out, tol = 0.01, burn_in = 1000)

Arguments

Value

Formatted liability estimate(s) and standard error(s) of the mean for the proband.

Examples

# uninformative sampling:
Gibbs_estimator(cov = diag(3), tbl = tibble::tibble(lower = rep(-Inf, 3),
upper = rep(Inf, 3)), out = 1:2, tol = 0.01, burn_in = 1000)

Title Pearson-Aitken algorithm to calculate mean values in truncated multivariate normal distributions

Description

Title Pearson-Aitken algorithm to calculate mean values in truncated multivariate normal distributions

Usage

PA_algorithm(mu, covmat, target_id, lower, upper, K_i = NA, K_pop = NA)

Arguments

Value

A list with two elements: est (expected genetic liability, given input data) and var (variance of genetic liability, given input data).

Attach attributes to a family graphs

Description

This function attaches attributes to family graphs, such as lower and upper thresholds, for each family member. This allows for a user-friendly way to attach personalised thresholds and other per-family specific attributes to the family graphs.

Usage

attach_attributes(
  cur_fam_graph,
  cur_proband,
  fid,
  attr_tbl,
  attr_names,
  censor_proband_thrs = TRUE
)

Arguments

Value

igraph object (neighbourhood graph around a proband) with updated attributes for each node in the graph.

Censor onset times in a family based on a proband's end of follow-up.

Description

This function censors onset times for family members based on the proband's end of follow-up. This is done to prevent using future events to base predictions on.

Usage

censor_family_onsets(
  tbl,
  proband_id_col,
  cur_proband,
  start,
  end,
  event,
  status_col = "status",
  aod_col = "aod",
  age_eof_col = "age"
)

Arguments

Value

tibble with updated end times, status, age of diagnosis, and age at end of follow-up for a family, such that proband's end time is used as the end time for all family members. This prevents using future events to based predictions on.

Examples

# See Vignettes.

Constructing a covariance matrix for a variable number of phenotypes

Description

construct_covmat returns the covariance matrix for an underlying target individual and a variable number of its family members for a variable number of phenotypes. It is a wrapper around construct_covmat_single and construct_covmat_multi.

Usage

construct_covmat(
  fam_vec = c("m", "f", "s1", "mgm", "mgf", "pgm", "pgf"),
  n_fam = NULL,
  add_ind = TRUE,
  h2 = 0.5,
  genetic_corrmat = NULL,
  full_corrmat = NULL,
  phen_names = NULL
)

Arguments

Details

This function can be used to construct a covariance matrix for a given number of family members. If h2 is a number, each entry in this covariance matrix equals the percentage of shared DNA between the corresponding individuals times the liability-scale heritability

h^2

. However, if h2 is a numeric vector, and genetic_corrmat and full_corrmat are two symmetric correlation matrices, each entry equals either the percentage of shared DNA between the corresponding individuals times the liability-scale heritability

h^2

or the percentage of shared DNA between the corresponding individuals times the correlation between the corresponding phenotypes. The family members can be specified using one of two possible formats.

Value

If either fam_vec or n_fam is used as the argument, if it is of the required format, if add_ind is a logical scalar and h2 is a number satisfying

0 \leq h2 \leq 1

, then the function construct_covmat will return a named covariance matrix, which row- and column-number corresponds to the length of fam_vec or n_fam (+ 2 if add_ind=TRUE). However, if h2 is a numeric vector satisfying

0 \leq h2_i \leq 1

for all

i \in \{1,...,n_pheno\}

and if genetic_corrmat and full_corrmat are two numeric and symmetric matrices satisfying that all diagonal entries are one and that all off-diagonal entries are between -1 and 1, then construct_covmat will return a named covariance matrix, which number of rows and columns corresponds to the number of phenotypes times the length of fam_vec or n_fam (+ 2 if add_ind=TRUE). If both fam_vec and n_fam are equal to c() or NULL, the function returns either a 2 \times 2 matrix holding only the correlation between the genetic component of the full liability and the full liability for the individual under consideration, or a

(2 \times n_pheno) \times (2\times n_pheno)

matrix holding the correlation between the genetic component of the full liability and the full liability for the underlying individual for all phenotypes. If both fam_vec and n_fam are specified, the user is asked to decide on which of the two vectors to use. Note that the returned object has different attributes, such as fam_vec, n_fam, add_ind and h2.

See Also

get_relatedness, construct_covmat_single, construct_covmat_multi

Examples

construct_covmat()
construct_covmat(fam_vec = c("m","mgm","mgf","mhs1","mhs2","mau1"),
                 n_fam = NULL,
                 add_ind = TRUE,
                 h2 = 0.5)
construct_covmat(fam_vec = NULL,
                 n_fam = stats::setNames(c(1,1,1,2,2), c("m","mgm","mgf","s","mhs")),
                 add_ind = FALSE,
                 h2 = 0.3)
construct_covmat(h2 = c(0.5,0.5), genetic_corrmat = matrix(c(1,0.4,0.4,1), nrow = 2),
                 full_corrmat = matrix(c(1,0.6,0.6,1), nrow = 2))

Constructing a covariance matrix for multiple phenotypes

Description

construct_covmat_multi returns the covariance matrix for an underlying target individual and a variable number of its family members for multiple phenotypes.

Usage

construct_covmat_multi(
  fam_vec = c("m", "f", "s1", "mgm", "mgf", "pgm", "pgf"),
  n_fam = NULL,
  add_ind = TRUE,
  genetic_corrmat,
  full_corrmat,
  h2_vec,
  phen_names = NULL
)

Arguments

Details

This function can be used to construct a covariance matrix for a given number of family members. Each entry in this covariance matrix equals either the percentage of shared DNA between the corresponding individuals times the liability-scale heritability h^2 or the percentage of shared DNA between the corresponding individuals times the correlation between the corresponding phenotypes. That is, for the same phenotype, the covariance between all combinations of the genetic component of the full liability and the full liability is given by

\text{Cov}\left( l_g, l_g \right) = h^2,

\text{Cov}\left( l_g, l_o \right) = h^2,

\text{Cov}\left( l_o, l_g \right) = h^2

and

\text{Cov}\left( l_o, l_o \right) = 1.

For two different phenotypes, the covariance is given by

\text{Cov}\left( l_g^1, l_g^2 \right) = \rho_g^{1,2},

\text{Cov}\left( l_g^1, l_o^2 \right) = \rho_g^{1,2},

\text{Cov}\left( l_o^1, l_g^2 \right) = \rho_g^{1,2}

and

\text{Cov}\left( l_o^1, l_o^2 \right) = \rho_g^{1,2} + \rho_e^{1,2},

where l_g^i and l_o^i are the genetic component of the full liability and the full liability for phenotype i, respectively, \rho_g^{i,j} is the genetic correlation between phenotype i and j and \rho_e^{1,2} is the environmental correlation between phenotype i and j. The family members can be specified using one of two possible formats.

Value

If either fam_vec or n_fam is used as the argument and if it is of the required format, if genetic_corrmat and full_corrmat are two numeric and symmetric matrices satisfying that all diagonal entries are one and that all off-diagonal entries are between -1 and 1, and if h2_vec is a numeric vector satisfying 0 \leq h2_i \leq 1 for all i \in \{1,...,n_pheno\}, then the output will be a named covariance matrix. The number of rows and columns corresponds to the number of phenotypes times the length of fam_vec or n_fam (+ 2 if add_ind=TRUE). If both fam_vec and n_fam are equal to c() or NULL, the function returns a (2 \times n_pheno) \times (2\times n_pheno) matrix holding only the correlation between the genetic component of the full liability and the full liability for the underlying individual for all phenotypes. If both fam_vec and n_fam are specified, the user is asked to decide on which of the two vectors to use. Note that the returned object has a number different attributes,namely fam_vec, n_fam, add_ind, genetic_corrmat, full_corrmat, h2 and phenotype_names.

See Also

get_relatedness, construct_covmat_single and construct_covmat.

Examples

construct_covmat_multi(fam_vec = NULL,
                       genetic_corrmat = matrix(c(1, 0.5, 0.5, 1), nrow = 2),
                       full_corrmat = matrix(c(1, 0.55, 0.55, 1), nrow = 2),
                       h2_vec = c(0.37,0.44),
                       phen_names = c("p1","p2"))
construct_covmat_multi(fam_vec = c("m","mgm","mgf","mhs1","mhs2","mau1"),
                       n_fam = NULL,
                       add_ind = TRUE,
                       genetic_corrmat = diag(3),
                       full_corrmat = diag(3),
                       h2_vec = c(0.8, 0.65))
construct_covmat_multi(fam_vec = NULL,
                       n_fam = stats::setNames(c(1,1,1,2,2), c("m","mgm","mgf","s","mhs")),
                       add_ind = FALSE,
                       genetic_corrmat = diag(2),
                       full_corrmat = diag(2),
                       h2_vec = c(0.75,0.85))

Constructing a covariance matrix for a single phenotype

Description

construct_covmatc_single returns the covariance matrix for an underlying target individual and a variable number of its family members

Usage

construct_covmat_single(
  fam_vec = c("m", "f", "s1", "mgm", "mgf", "pgm", "pgf"),
  n_fam = NULL,
  add_ind = TRUE,
  h2 = 0.5
)

Arguments

Details

This function can be used to construct a covariance matrix for a given number of family members. Each entry in this covariance matrix equals the percentage of shared DNA between the corresponding individuals times the liability-scale heritability h^2. The family members can be specified using one of two possible formats.

Value

If either fam_vec or n_fam is used as the argument, if it is of the required format and h2 is a number satisfying 0 \leq h2 \leq 1, then the output will be a named covariance matrix. The number of rows and columns corresponds to the length of fam_vec or n_fam (+ 2 if add_ind=TRUE). If both fam_vec = c()/NULL and n_fam = c()/NULL, the function returns a 2 \times 2 matrix holding only the correlation between the genetic component of the full liability and the full liability for the individual. If both fam_vec and n_fam are given, the user is asked to decide on which of the two vectors to use. Note that the returned object has different attributes, such as fam_vec, n_fam, add_ind and h2.

See Also

get_relatedness, construct_covmat_multi, construct_covmat

Examples

construct_covmat_single()
construct_covmat_single(fam_vec = c("m","mgm","mgf","mhs1","mhs2","mau1"),
n_fam = NULL, add_ind = TRUE, h2 = 0.5)
construct_covmat_single(fam_vec = NULL, n_fam = stats::setNames(c(1,1,1,2,2),
c("m","mgm","mgf","s","mhs")), add_ind = FALSE, h2 = 0.3)

Convert age to cumulative incidence rate

Description

convert_age_to_cir computes the cumulative incidence rate from a person's age.

Usage

convert_age_to_cir(age, pop_prev = 0.1, mid_point = 60, slope = 1/8)

Arguments

Details

Given a person's age, convert_age_to_cir can be used to compute the cumulative incidence rate (cir), which is given by the formula

pop\_ prev / (1 + exp((mid\_ point - age) * slope))

Value

If age and mid_point are positive numbers, if pop_prev is a positive number between 0 and 1 and if slope is a valid number, then convert_age_to_cir returns a number, which is equal to the cumulative incidence rate.

Examples

curve(sapply(age, convert_age_to_cir), from = 10, to = 110, xname = "age")

Convert age to threshold

Description

convert_age_to_thresh computes the threshold from a person's age using either the logistic function or the truncated normal distribution

Usage

convert_age_to_thresh(
  age,
  dist = "logistic",
  pop_prev = 0.1,
  mid_point = 60,
  slope = 1/8,
  min_age = 10,
  max_age = 90,
  lower = stats::qnorm(0.05, lower.tail = FALSE),
  upper = Inf
)

Arguments

Details

Given a person's age, convert_age_to_thresh can be used to first compute the cumulative incidence rate (cir), which is then used to compute the threshold using either the logistic function or the truncated normal distribution. Under the logistic function, the formula used to compute the threshold from an individual's age is given by

qnorm(pop\_ prev / (1 + exp((mid\_ point - age) * slope)), lower.tail = F)

, while it is given by

qnorm((1 - (age-min\_ age)/max\_ age) * (pnorm(upper) - pnorm(lower)) + pnorm(lower))

under the truncated normal distribution.

Value

If age is a positive number and all other necessary arguments are valid, then convert_age_to_thresh returns a number, which is equal to the threshold.

Examples

curve(sapply(age, convert_age_to_thresh), from = 10, to = 110, xname = "age")

Convert cumulative incidence rate to age

Description

convert_cir_to_age computes the age from a person's cumulative incidence rate.

Usage

convert_cir_to_age(cir, pop_prev = 0.1, mid_point = 60, slope = 1/8)

Arguments

Details

Given a person's cumulative incidence rate (cir), convert_cir_to_age can be used to compute the corresponding age, which is given by

mid\_ point - \log(pop\_ prev/cir - 1) * 1/slope

Value

If cir and mid_point are positive numbers, if pop_prev is a positive number between 0 and 1 and if slope is a valid number, then convert_cir_to_age returns a number, which is equal to the current age.

Examples

curve(sapply(cir, convert_cir_to_age), from = 0.001, to = 0.099, xname = "cir")

Attempts to convert the list entry input format to a long format

Description

Attempts to convert the list entry input format to a long format

Usage

convert_format(family, threshs, personal_id_col = "pid", role_col = NULL)

Arguments

Value

returns a format similar to prepare_thresholds, which is used by estimate_liability

Examples

family <- data.frame(
fid = c(1, 1, 1, 1),
pid = c(1, 2, 3, 4),
role = c("o", "m", "f", "pgf")
)

threshs <- data.frame(
  pid = c(1, 2, 3, 4),
  lower = c(-Inf, -Inf, 0.8, 0.7),
  upper = c(0.8, 0.8, 0.8, 0.7)
)

convert_format(family, threshs)

Convert liability to age of onset

Description

convert_liability_to_aoo computes the age of onset from an individual's true underlying liability using either the logistic function or the truncated normal distribution.

Usage

convert_liability_to_aoo(
  liability,
  dist = "logistic",
  pop_prev = 0.1,
  mid_point = 60,
  slope = 1/8,
  min_aoo = 10,
  max_aoo = 90,
  lower = stats::qnorm(0.05, lower.tail = FALSE),
  upper = Inf
)

Arguments

Details

Given a person's cumulative incidence rate (cir), convert_liability_to_aoo can be used to compute the corresponding age. Under the logistic function, the age is given by

mid\_ point - log(pop\_ prev/cir - 1) * 1/slope

, while it is given by

(1 - truncated\_ normal\_ cdf(liability = liability, lower = lower , upper = upper)) * max\_ aoo + min\_ aoo

under the truncated normal distribution.

Value

If liability is a number and all other necessary arguments are valid, then convert_liability_to_aoo returns a positive number, which is equal to the age of onset.

Examples

curve(sapply(liability, convert_liability_to_aoo), from = 1.3, to = 3.5, xname = "liability")
curve(sapply(liability, convert_liability_to_aoo, dist = "normal"),
 from = qnorm(0.05, lower.tail = FALSE), to = 3.5, xname = "liability")

Convert the heritability on the observed scale to that on the liability scale

Description

convert_observed_to_liability_scale transforms the heritability on the observed scale to the heritability on the liability scale.

Usage

convert_observed_to_liability_scale(
  obs_h2 = 0.5,
  pop_prev = 0.05,
  prop_cases = 0.5
)

Arguments

Details

This function can be used to transform the heritability on the observed scale to that on the liability scale. convert_observed_to_liability_scale uses either Equation 17 (if prop_cases = NULL) or Equation 23 from Sang Hong Lee, Naomi R. Wray, Michael E. Goddard and Peter M. Visscher, "Estimating Missing Heritability for Diseases from Genome-wide Association Studies", The American Journal of Human Genetics, Volume 88, Issue 3, 2011, pp. 294-305, doi:10.1016/j.ajhg.2011.02.002 to transform the heritability on the observed scale to the heritability on the liability scale.

Value

If obs_h2, pop_prev and prop_cases are non-negative numbers that are at most one, the function returns the heritability on the liability scale using Equation 23 from Sang Hong Lee, Naomi R. Wray, Michael E. Goddard and Peter M. Visscher, "Estimating Missing Heritability for Diseases from Genome-wide Association Studies", The American Journal of Human Genetics, Volume 88, Issue 3, 2011, pp. 294-305, doi:10.1016/j.ajhg.2011.02.002. If obs_h2, pop_prev and prop_cases are non-negative numeric vectors where all entries are at most one, the function returns a vector of the same length as obs_h2. Each entry holds to the heritability on the liability scale which was obtained from the corresponding entry in obs_h2 using Equation 23. If obs_h2 and pop_prev are non-negative numbers that are at most one and prop_cases is NULL, the function returns the heritability on the liability scale using Equation 17 from Sang Hong Lee, Naomi R. Wray, Michael E. Goddard and Peter M. Visscher, "Estimating Missing Heritability for Diseases from Genome-wide Association Studies", The American Journal of Human Genetics, Volume 88, Issue 3, 2011, pp. 294-305, doi:10.1016/j.ajhg.2011.02.002. If obs_h2 and pop_prev are non-negative numeric vectors such that all entries are at most one, while prop_cases is NULL, convert_observed_to_liability_scale returns a vector of the same length as obq_h2. Each entry holds to the liability-scale heritability that was obtained from the corresponding entry in obs_h2 using Equation 17.

References

Sang Hong Lee, Naomi R. Wray, Michael E. Goddard, Peter M. Visscher (2011, March). Estimating Missing Heritability for Diseases from Genome-wide Association Studies. In The American Journal of Human Genetics (Vol. 88, Issue 3, pp. 294-305). doi:10.1016/j.ajhg.2011.02.002

Examples

convert_observed_to_liability_scale()
convert_observed_to_liability_scale(prop_cases=NULL)
convert_observed_to_liability_scale(obs_h2 = 0.8, pop_prev = 1/44,
                                    prop_cases = NULL)
convert_observed_to_liability_scale(obs_h2 = c(0.5,0.8),
                                    pop_prev = c(0.05, 1/44),
                                    prop_cases = NULL)

Positive definite matrices

Description

correct_positive_definite verifies that a given covariance matrix is indeed positive definite by checking that all eigenvalues are positive. If the given covariance matrix is not positive definite, correct_positive_definite tries to modify the underlying correlation matrices genetic_corrmat and full_corrmat in order to obtain a positive definite covariance matrix.

Usage

correct_positive_definite(
  covmat,
  correction_val = 0.99,
  correction_limit = 100
)

Arguments

Details

This function can be used to verify that a given covariance matrix is positive definite. It calculates all eigenvalues in order to investigate whether they are all positive. This property is necessary for the covariance matrix to be used as a Gaussian covariance matrix. It is especially useful to check whether any covariance matrix obtained by construct_covmat_multi is positive definite. If the given covariance matrix is not positive definite, correct_positive_definite tries to modify the underlying correlation matrices (called genetic_corrmat and full_corrmat in construct_covmat or construct_covmat_multi) by multiplying all off-diagonal entries in the correlation matrices by a given number.

Value

If covmat is a symmetric and numeric matrix and all eigenvalues are positive, correct_positive_definite simply returns covmat. If some eigenvalues are not positive and correction_val is a positive number, correct_positive_definite tries to convert covmat into a positive definite matrix. If covmat has attributes add_ind, h2, genetic_corrmat, full_corrmat and phenotype_names, correct_positive_definite computes a new covariance matrix using slightly modified correlation matrices genetic_corrmat and full_corrmat. If the correction is performed successfully, i.e. if the new covariance matrix is positive definite,the new covariance matrix is returned. Otherwise, correct_positive_definite returns the original covariance matrix.

See Also

construct_covmat, construct_covmat_single and construct_covmat_multi.

Examples

ntrait <- 2
genetic_corrmat <- matrix(0.6, ncol = ntrait, nrow = ntrait)
diag(genetic_corrmat) <- 1
full_corrmat <- matrix(-0.25, ncol = ntrait, nrow = ntrait)
diag(full_corrmat) <- 1
h2_vec <- rep(0.6, ntrait)
cov <- construct_covmat(fam_vec = c("m", "f"),
  genetic_corrmat = genetic_corrmat,
  h2 = h2_vec,
  full_corrmat = full_corrmat)
cov
correct_positive_definite(cov)

Estimate genetic liability similar to LT-FH

Description

Estimate genetic liability similar to LT-FH

Usage

estimate_gen_liability_ltfh(
  h2,
  phen,
  child_threshold,
  parent_threshold,
  status_col_offspring = "CHILD_STATUS",
  status_col_father = "P1_STATUS",
  status_col_mother = "P2_STATUS",
  status_col_siblings = "SIB_STATUS",
  number_of_siblings_col = "NUM_SIBS",
  tol = 0.01
)

Arguments

Value

Returns the estimated genetic liabilities.

Examples

phen <- data.frame(
CHILD_STATUS = c(0,0),
P1_STATUS = c(1,1),
P2_STATUS = c(0,1),
SIB_STATUS = c(1,0),
NUM_SIBS = c(2,0))

h2 <- 0.5
child_threshold <- 0.7
parent_threshold <- 0.8

estimate_gen_liability_ltfh(h2, phen, child_threshold, parent_threshold)

Estimating the genetic or full liability for a variable number of phenotypes

Description

estimate_liability estimates the genetic component of the full liability and/or the full liability for a number of individuals based on their family history for one or more phenotypes. It is a wrapper around estimate_liability_single and estimate_liability_multi.

Usage

estimate_liability(
  .tbl = NULL,
  family_graphs = NULL,
  h2 = 0.5,
  pid = "pid",
  fid = "fid",
  role = "role",
  family_graphs_col = "fam_graph",
  out = c(1),
  tol = 0.01,
  method = "PA",
  useMixture = FALSE,
  genetic_corrmat = NULL,
  full_corrmat = NULL,
  phen_names = NULL
)

Arguments

Details

This function can be used to estimate either the genetic component of the full liability, the full liability or both for a variable number of traits.

Value

If family and threshs are two matrices, lists or data frames that can be converted into tibbles, if family has two columns named like the strings represented in pid and fid, if threshs has a column named like the string given in pid as well as a column named "lower" and a column named "upper" and if the liability-scale heritability h2 is a number (length(h2)=1), and out, tol and always_add are of the required form, then the function returns a tibble with either four or six columns (depending on the length of out). The first two columns correspond to the columns fid and pid ' present in family. If out is equal to c(1) or c("genetic"), the third and fourth column hold the estimated genetic liability as well as the corresponding standard error, respectively. If out equals c(2) or c("full"), the third and fourth column hold the estimated full liability as well as the corresponding standard error, respectively. If out is equal to c(1,2) or c("genetic","full"), the third and fourth column hold the estimated genetic liability as well as the corresponding standard error, respectively, while the fifth and sixth column hold the estimated full liability as well as the corresponding standard error, respectively. If h2 is a numeric vector of length greater than 1 and if genetic_corrmat, full_corrmat, out and tol are of the required form, then the function returns a tibble with at least six columns (depending on the length of out). The first two columns correspond to the columns fid and pid present in the tibble family. If out is equal to c(1) or c("genetic"), the third and fourth columns hold the estimated genetic liability as well as the corresponding standard error for the first phenotype, respectively. If out equals c(2) or c("full"), the third and fourth columns hold the estimated full liability as well as the corresponding standard error for the first phenotype, respectively. If out is equal to c(1,2) or c("genetic","full"), the third and fourth columns hold the estimated genetic liability as well as the corresponding standard error for the first phenotype, respectively, while the fifth and sixth columns hold the estimated full liability as well as the corresponding standard error for the first phenotype, respectively. The remaining columns hold the estimated genetic liabilities and/or the estimated full liabilities as well as the corresponding standard errors for the remaining phenotypes.

See Also

future_apply, estimate_liability_single, estimate_liability_multi

Examples

genetic_corrmat <- matrix(0.4, 3, 3)
diag(genetic_corrmat) <- 1
full_corrmat <- matrix(0.6, 3, 3)
diag(full_corrmat) <- 1
#
sims <- simulate_under_LTM(fam_vec = c("m","f"), n_fam = NULL, add_ind = TRUE,
genetic_corrmat = genetic_corrmat, full_corrmat = full_corrmat, h2 = rep(.5,3),
n_sim = 1, pop_prev = rep(.1,3))
estimate_liability(.tbl = sims$thresholds, h2 = rep(.5,3),
genetic_corrmat = genetic_corrmat, full_corrmat = full_corrmat,
pid = "indiv_ID", fid = "fid", role = "role", out = c(1),
phen_names = paste0("phenotype", 1:3), tol = 0.01)

Estimating the genetic or full liability for multiple phenotypes

Description

estimate_liability_multi estimates the genetic component of the full liability and/or the full liability for a number of individuals based on their family history for a variable number of phenotypes.

Usage

estimate_liability_multi(
  .tbl = NULL,
  family_graphs = NULL,
  h2_vec,
  genetic_corrmat,
  full_corrmat,
  phen_names = NULL,
  pid = "pid",
  fid = "fid",
  role = "role",
  family_graphs_col = "fam_graph",
  out = c(1),
  tol = 0.01
)

Arguments

Details

This function can be used to estimate either the genetic component of the full liability, the full liability or both for a variable number of traits.

Value

If family and threshs are two matrices, lists or data frames that can be converted into tibbles, if family has two columns named like the strings represented in pid and fid, if threshs has a column named like the string given in pid as well as a column named "lower" and a column named "upper" and if the liability-scale heritabilities in h2_vec, genetic_corrmat, full_corrmat, out and tol are of the required form, then the function returns a tibble with at least six columns (depending on the length of out). The first two columns correspond to the columns fid and pid present in the tibble family. If out is equal to c(1) or c("genetic"), the third and fourth columns hold the estimated genetic liability as well as the corresponding standard error for the first phenotype, respectively. If out equals c(2) or c("full"), the third and fourth columns hold the estimated full liability as well as the corresponding standard error for the first phenotype, respectively. If out is equal to c(1,2) or c("genetic","full"), the third and fourth columns hold the estimated genetic liability as well as the corresponding standard error for the first phenotype, respectively, while the fifth and sixth columns hold the estimated full liability as well as the corresponding standard error for the first phenotype, respectively. The remaining columns hold the estimated genetic liabilities and/or the estimated full liabilities as well as the corresponding standard errors for the remaining phenotypes.

See Also

future_apply, estimate_liability_single, estimate_liability

Examples

genetic_corrmat <- matrix(0.4, 3, 3)
diag(genetic_corrmat) <- 1
full_corrmat <- matrix(0.6, 3, 3)
diag(full_corrmat) <- 1
#
sims <- simulate_under_LTM(fam_vec = c("m","f"), n_fam = NULL, add_ind = TRUE,
genetic_corrmat = genetic_corrmat, full_corrmat = full_corrmat, h2 = rep(.5,3),
n_sim = 1, pop_prev = rep(.1,3))
estimate_liability_multi(.tbl = sims$thresholds, h2_vec = rep(.5,3),
genetic_corrmat = genetic_corrmat, full_corrmat = full_corrmat,
pid = "indiv_ID", fid = "fid", role = "role", out = c(1),
phen_names = paste0("phenotype", 1:3), tol = 0.01)

Estimating the genetic or full liability

Description

estimate_liability_single estimates the genetic component of the full liability and/or the full liability for a number of individuals based on their family history.

Usage

estimate_liability_single(
  .tbl = NULL,
  family_graphs = NULL,
  h2 = 0.5,
  pid = "pid",
  fid = "fid",
  family_graphs_col = "fam_graph",
  role = NULL,
  out = c(1),
  tol = 0.01,
  useMixture = FALSE,
  method = "PA"
)

Arguments

Details

This function can be used to estimate either the genetic component of the full liability, the full liability or both. It is possible to input either

Value

If family and threshs are two matrices, lists or data frames that can be converted into tibbles, if family has two columns named like the strings represented in pid and fid, if threshs has a column named like the string given in pid as well as a column named "lower" and a column named "upper" and if the liability-scale heritability h2, out, tol and always_add are of the required form, then the function returns a tibble with either four or six columns (depending on the length of out). The first two columns correspond to the columns fid and pid ' present in family. If out is equal to c(1) or c("genetic"), the third and fourth column hold the estimated genetic liability as well as the corresponding standard error, respectively. If out equals c(2) or c("full"), the third and fourth column hold the estimated full liability as well as the corresponding standard error, respectively. If out is equal to c(1,2) or c("genetic","full"), the third and fourth column hold the estimated genetic liability as well as the corresponding standard error, respectively, while the fifth and sixth column hold the estimated full liability as well as the corresponding standard error, respectively.

See Also

future_apply, estimate_liability_multi, estimate_liability

Examples

sims <- simulate_under_LTM(fam_vec = c("m","f","s1"), n_fam = NULL,
add_ind = TRUE, h2 = 0.5, n_sim=10, pop_prev = .05)
#
estimate_liability_single(.tbl = sims$thresholds,
h2 = 0.5, pid = "indiv_ID", fid = "fid", role = "role", out = c(1),
tol = 0.01)
#
sims <- simulate_under_LTM(fam_vec = c(), n_fam = NULL, add_ind = TRUE,
h2 = 0.5, n_sim=10, pop_prev = .05)
#
estimate_liability_single(.tbl = sims$thresholds,
h2 = 0.5, pid = "indiv_ID", fid = "fid", role = "role",
out = c("genetic"), tol = 0.01)

Title Internal Function used to extact input needed from graph input for liability estimation

Description

Title Internal Function used to extact input needed from graph input for liability estimation

Usage

extract_estimation_info_graph(cur_fam_graph, cur_fid, h2, pid, add_ind = TRUE)

Arguments

Value

list with two elements: tbl (tibble with all relevant information) and cov (covariance matrix) estimated through graph_based_covariance_construction()

Title Internal Function used to extact input needed for liability estimation

Description

Title Internal Function used to extact input needed for liability estimation

Usage

extract_estimation_info_tbl(.tbl, cur_fid, h2, fid, pid, role, add_ind = TRUE)

Arguments

Value

list with two elements: tbl (tibble with all relevant information) and cov (covariance matrix) estimated through construct_covmat()

Wrapper to attach attributes to family graphs

Description

This function can attach attributes to family graphs, such as lower and upper thresholds, for each family member. This allows for personalised thresholds and other per-family specific attributes. This function wraps around attach_attributes to ease the process of attaching attributes to family graphs in the standard format.

Usage

familywise_attach_attributes(
  family_graphs,
  fam_attr,
  fam_graph_col = "fam_graph",
  attached_fam_graph_col = "masked_fam_graph",
  fid = "fid",
  pid = "pid",
  cols_to_attach = c("lower", "upper"),
  censor_proband_thrs = TRUE
)

Arguments

Value

tibble with family ids and an updated family graph with attached attributes. If lower and upper thresholds are specified, the input is ready for estimate_liability().

Examples

# See Vignettes.

Censor Family Onsets for Multiple Families

Description

This fucntion is a wrapper around censor_family_onsets. This functions accepts a tibble with family graphs from get_family_graphs. It censors the onset times for each individual in the family graph based on the proband's end of follow-up. Returns a formatted output.

Usage

familywise_censoring(
  family_graphs,
  tbl,
  start,
  end,
  event,
  status_col = "status",
  aod_col = "aod",
  age_eof_col = "age",
  fam_graph_col = "fam_graph",
  fid = "fid",
  pid = "pid",
  merge_by = pid
)

Arguments

Value

A tibble with family ids and updated status, age of diagnosis, and age at end of follow-up for each individual in the family based on the proband's end of follow-up.

Examples

# See Vignettes.

Fixing sex coding in trio info

Description

Internal function used to assist in fixing sex coding separately from id coding type.

Usage

fixSexCoding(x, sex_coding = TRUE, dadid, momid)

Arguments

Value

appropriate sex coding

construct all combinations of input vector

Description

pastes together all combinations of input vector

Usage

get_all_combs(vec)

Arguments

Value

A vector of strings is returned.

Examples

get_all_combs(letters[1:3])

Construct kinship matrix from graph

Description

construct the kinship matrix from a graph representation of a family, centered on an index person (proband).

Usage

get_covmat(fam_graph, h2, index_id = NA, add_ind = TRUE, fix_diag = TRUE)

Arguments

Value

A kinship matrix.

Examples

fam <- data.frame(
i = c(1, 2, 3, 4),
f = c(3, 0, 4, 0),
m = c(2, 0, 0, 0)
)

thresholds <- data.frame(
  i = c(1, 2, 3, 4),
  lower = c(-Inf, -Inf, 0.8, 0.7),
  upper = c(0.8, 0.8, 0.8, 0.7)
)

graph <- prepare_graph(fam, icol = "i", fcol = "f", mcol = "m", node_attributes = thresholds)

get_covmat(graph, h2 = 0.5, index_id = "1")

Automatically identify family members of degree n

Description

This function identifies individuals ndegree-steps away from the proband in the population graph.

Usage

get_family_graphs(
  pop_graph,
  ndegree,
  proband_vec,
  fid = "fid",
  fam_graph_col = "fam_graph",
  mindist = 0,
  mode = "all"
)

Arguments

Value

Tibble with two columns, family ids (fid) and family graphs (fam_graph_col).

Examples

# See Vignettes.

Calculate age of diagnosis, age at end of follow up, and status

Description

Calculate age of diagnosis, age at end of follow up, and status

Usage

get_onset_time(
  tbl,
  start,
  end,
  event,
  status_col = "status",
  aod_col = "aod",
  age_eof_col = "age"
)

Arguments

Value

tibble with added status, age of diagnosis, and age at end of follow-up

Examples

# See vignettes.

Relatedness between a pair of family members

Description

get_relatedness returns the relatedness times the liability-scale heritability for a pair of family members

Usage

get_relatedness(s1, s2, h2 = 0.5, from_covmat = FALSE)

Arguments

Details

This function can be used to get the percentage of shared DNA times the liability-scale heritability h^2 for two family members.

Value

If both s1 and s2 are strings chosen from the mentioned list of strings and h2 is a number satisfying 0 \leq h2 \leq 1, then the output will be a number that equals the percentage of shared DNA between s1 and s2 times the squared heritability h2.

Note

If you are only interested in the percentage of shared DNA, set h2 = 1.

Examples

get_relatedness("g","o")
get_relatedness("g","f", h2 = 1)
get_relatedness("o","s", h2 = 0.3)


# This will result in errors:
try(get_relatedness("a","b"))
try(get_relatedness(m, mhs))

Constructing covariance matrix from local family graph

Description

Function that constructs the genetic covariance matrix given a graph around a proband and extracts the threshold information from the graph.

Usage

graph_based_covariance_construction(
  pid,
  cur_proband_id,
  cur_family_graph,
  h2,
  add_ind = TRUE
)

Arguments

Value

list with two elements. The first element is temp_tbl, which contains the id of the current proband, the family ID and the lower and upper thresholds. The second element, cov, is the covariance matrix of the local graph centered on the current proband.

Examples

fam <- data.frame(
  id = c("pid", "mom", "dad", "pgf"),
  dadcol = c("dad", 0, "pgf", 0),
  momcol = c("mom", 0, 0, 0))

thresholds <- data.frame(
  id = c("pid", "mom", "dad", "pgf"),
  lower = c(-Inf, -Inf, 0.8, 0.7),
  upper = c(0.8, 0.8, 0.8, 0.7))

graph <- prepare_graph(fam, icol = "id", fcol = "dadcol",
 mcol = "momcol", node_attributes = thresholds)

graph_based_covariance_construction(pid = "id",
                                    cur_proband_id = "pid",
                                    cur_family_graph = graph,
                                    h2 = 0.5)

Constructing covariance matrix from local family graph for multi trait analysis

Description

Function that constructs the genetic covariance matrix given a graph around a proband and extracts the threshold information from the graph.

Usage

graph_based_covariance_construction_multi(
  fid,
  pid,
  cur_proband_id,
  cur_family_graph,
  h2_vec,
  genetic_corrmat,
  phen_names,
  add_ind = TRUE
)

Arguments

Value

list with three elements. The first element is temp_tbl, which contains the id of the current proband, the family ID and the lower and upper thresholds for all phenotypes. The second element, cov, is the covariance matrix of the local graph centred on the current proband. The third element is newOrder, which is the order of ids from pid and phen_names pasted together, such that order can be enforced elsewhere too.

Examples

fam <- data.frame(
fam = c(1, 1, 1,1),
id = c("pid", "mom", "dad", "pgf"),
dadcol = c("dad", 0, "pgf", 0),
momcol = c("mom", 0, 0, 0))

thresholds <- data.frame(
  id = c("pid", "mom", "dad", "pgf"),
  lower_1 = c(-Inf, -Inf, 0.8, 0.7),
  upper_1 = c(0.8, 0.8, 0.8, 0.7),
  lower_2 = c(-Inf, 0.3, -Inf, 0.2),
  upper_2 = c(0.3, 0.3, 0.3, 0.2))

graph <- prepare_graph(fam, icol = "id", fcol = "dadcol",
 mcol = "momcol", node_attributes = thresholds)

ntrait <- 2
genetic_corrmat <- matrix(0.2, ncol = ntrait, nrow = ntrait)
diag(genetic_corrmat) <- 1
full_corrmat <- matrix(0.3, ncol = ntrait, nrow = ntrait)
diag(full_corrmat) <- 1
h2_vec <- rep(0.6, ntrait)

graph_based_covariance_construction_multi(fid = "fam",
                                          pid = "id",
                                          cur_proband_id = "pid",
                                          cur_family_graph = graph,
                                          h2_vec = h2_vec,
                                          genetic_corrmat = genetic_corrmat,
                                          phen_names = c("1", "2"))

Convert from igraph to trio information

Description

This function converts an igraph object to a trio information format.

Usage

graph_to_trio(
  graph,
  id = "id",
  dadid = "dadid",
  momid = "momid",
  sex = "sex",
  fixParents = TRUE
)

Arguments

Details

The sex column is required in the igraph attributes. The sex information is used to determine who is the mother and father in the trio.

Value

A tibble with trio information.

Examples

if (FALSE) {

family = tribble(
~id, ~momcol, ~dadcol,
"pid", "mom", "dad",
"sib", "mom", "dad",
"mhs", "mom", "dad2",
"phs", "mom2", "dad",
"mom", "mgm", "mgf",
"dad", "pgm", "pgf",
"dad2", "pgm2", "pgf2",
"paunt", "pgm", "pgf",
"pacousin", "paunt", "pauntH",
"hspaunt", "pgm", "newpgf",
"hspacousin", "hspaunt", "hspauntH",
"puncle", "pgm", "pgf",
"pucousin", "puncleW", "puncle",
"maunt", "mgm", "mgf",
"macousin", "maunt", "mauntH",
"hsmuncle", "newmgm", "mgf",
"hsmucousin", "hsmuncleW", "hsmuncle"
)


thrs =  tibble(
  id = family %>% select(1:3) %>% unlist() %>% unique(),
  lower = sample(c(-Inf, 2), size = length(id), replace = TRUE),
  upper = sample(c(2, Inf), size = length(id), replace = TRUE),
 sex = case_when(
   id %in% family$momcol ~ "F",
    id %in% family$dadcol ~ "M",
    TRUE ~ NA)) %>%
  mutate(sex = sapply(sex, function(x) ifelse(is.na(x),
  sample(c("M", "F"), 1), x)))
graph = prepare_graph(.tbl = family,
icol = "id", fcol = "dadcol", mcol = "momcol", node_attributes = thrs)
}

Title Kendler's FGRS

Description

Title Kendler's FGRS

Usage

kendler(
  .tbl = NULL,
  family_graphs = NULL,
  family_graphs_col = "fam_graph",
  pid = "pid",
  fid = "fid",
  role = NULL,
  dadcol,
  momcol,
  env_cor_sib = 1,
  env_cor_f = 1,
  env_cor_m = 1
)

Arguments

Value

A tibble with summary values used to calculate the kendler FGRS and the FGRS itself.

Examples

# See Vignettes.

Title Helper function for Kendler's FGRS

Description

Title Helper function for Kendler's FGRS

Usage

kendler_family_calculations(
  tbl,
  cov,
  pid,
  cur_dad_id,
  cur_mom_id,
  env_cor_sib = 1,
  env_cor_f = 1,
  env_cor_m = 1
)

Arguments

Value

A tibble with family specific values required for Kendler's FGRS calculation.

Examples

# See Vignettes.

Construct graph from register information

Description

prepare_graph constructs a graph based on mother, father, and offspring links.

Usage

prepare_graph(
  .tbl,
  icol,
  fcol,
  mcol,
  node_attributes = NA,
  missingID_patterns = "^0$"
)

Arguments

Value

An igraph object. A (directed) graph object based on the links provided in .tbl, potentially with provided attributes stored for each node.

Examples

fam <- data.frame(
  id = c("pid", "mom", "dad", "pgf"),
  dadcol = c("dad", 0, "pgf", 0),
  momcol = c("mom", 0, 0, 0))

thresholds <- data.frame(
  id = c("pid", "mom", "dad", "pgf"),
  lower = c(-Inf, -Inf, 0.8, 0.7),
  upper = c(0.8, 0.8, 0.8, 0.7))

prepare_graph(fam, icol = "id", fcol = "dadcol", mcol = "momcol", node_attributes = thresholds)

Calculate (personalised) thresholds based on CIPs.

Description

This function prepares input for estimate_liability by calculating thresholds based on stratified cumulative incidence proportions (CIPs) with options for interpolation for ages between CIP values. Given a tibble with families and family members and (stratified) CIPs, personalised thresholds will be calculated for each individual present in .tbl. An individual may be in multiple families, but only once in the same family.

Usage

prepare_thresholds(
  .tbl,
  CIP,
  age_col,
  CIP_merge_columns = c("sex", "birth_year", "age"),
  CIP_cip_col = "cip",
  Kpop = "useMax",
  status_col = "status",
  lower_equal_upper = FALSE,
  personal_thr = FALSE,
  fid_col = "fid",
  personal_id_col = "pid",
  interpolation = NULL,
  bst.params = list(max_depth = 10, base_score = 0, nthread = 4, min_child_weight = 10),
  min_CIP_value = 1e-05,
  xgboost_itr = 30
)

Arguments

Value

Tibble with (personlised) thresholds for each family member (lower & upper), the calculated cumulative incidence proportion for each individual (K_i), and population prevalence within an individuals CIP strata (K_pop; max value in stratum). The threshold and other potentially relevant information can be added to the family graphs with familywise_attach_attributes.

Examples

tbl = data.frame(
fid = c(1, 1, 1, 1),
pid = c(1, 2, 3, 4),
role = c("o", "m", "f", "pgf"),
sex = c(1, 0, 1, 1),
status = c(0, 0, 1, 1),
age = c(22, 42, 48, 78),
birth_year = 2023 - c(22, 42, 48, 78),
aoo = c(NA, NA, 43, 45))

cip = data.frame(
age = c(22, 42, 43, 45, 48, 78),
birth_year = c(2001, 1981, 1975, 1945, 1975, 1945),
sex = c(1, 0, 1, 1, 1, 1),
cip = c(0.1, 0.2, 0.3, 0.3, 0.3, 0.4))

prepare_thresholds(.tbl = tbl, CIP = cip, age_col = "age", interpolation = NA)

Gibbs Sampler for the truncated multivariate normal distribution

Description

rtmvnorm.gibbs implements Gibbs sampler for the truncated multivariate normal distribution with covariance matrix covmat.

Usage

rtmvnorm.gibbs(
  n_sim = 1e+05,
  covmat,
  lower = -Inf,
  upper,
  fixed = (lower == upper),
  out = c(1),
  burn_in = 1000
)

Arguments

Details

Given a covariance matrix covmat and lower and upper cutoff points, the function rtmvnorm.gibbs() can be used to perform Gibbs sampler on a truncated multivariable normal distribution. It is possible to specify which variables to return from the Gibbs sampler, making it convenient to use when estimating only the full liability or the genetic component of the full liability.

Value

If covmat is a symmetric and numeric matrix, if n_sim and burn_in are positive/non-negative numbers, if out is a numeric vector and lower, upper and fixed are numbers or vectors of the same length and the required format, rtmvnorm.gibbs returns the sampling values from the Gibbs sampler for all variables specified in out.

References

Kotecha, J. H., & Djuric, P. M. (1999, March). Gibbs sampling approach for generation of truncated multivariate gaussian random variables. In 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No. 99CH36258) (Vol. 3, pp. 1757-1760). IEEE. doi:10.1109/ICASSP.1999.756335

Wilhelm, S., & Manjunath, B. G. (2010). tmvtnorm: A package for the truncated multivariate normal distribution. The R Journal. doi:10.32614/RJ-2010-005

Examples

samp <- rtmvnorm.gibbs(10e3, covmat = matrix(c(1, 0.2, 0.2, 0.5), 2),
                       lower = c(-Inf, 0), upper = c(0, Inf), out = 1:2)

Simulate under the liability threshold model.

Description

simulate_under_LTM simulates families and thresholds under the liability threshold model for a given family structure and a variable number of phenotypes.Please note that it is not possible to simulate different family structures.

Usage

simulate_under_LTM(
  fam_vec = c("m", "f", "s1", "mgm", "mgf", "pgm", "pgf"),
  n_fam = NULL,
  add_ind = TRUE,
  h2 = 0.5,
  genetic_corrmat = NULL,
  full_corrmat = NULL,
  phen_names = NULL,
  n_sim = 1000,
  pop_prev = 0.1
)

Arguments

Details

This function can be used to simulate the case-control status, the current age and age-of-onset as well as the lower and upper thresholds for a variable number of phenotypes for all family members in each of the n_sim families. If h2 is a number, simulate_under_LTM simulates the case- control status, the current age and age-of-onset as well as thresholds for a single phenotype. However, if h2 is a numeric vector, if genetic_corrmat and full_corrmat are two symmetric correlation matrices, and if phen_names and pop_prev are to numeric vectors holding the phenotype names and the population prevalences, respectively, then simulate_under_LTM simulates the case-control status, the current age and age-of-onset as well as thresholds for two or more (correlated) phenotypes. The family members can be specified using one of two possible formats.

Value

If either fam_vec or n_fam is used as the argument, if it is of the required format, if the liability-scale heritability h2 is a number satisfying 0 \leq h^2, n_sim is a strictly positive number, and pop_prev is a positive number that is at most one, then the output will be a list containing two tibbles. The first tibble, sim_obs, holds the simulated liabilities, the disease status and the current age/age-of-onset for all family members in each of the n_sim families. The second tibble, thresholds, holds the family identifier, the personal identifier, the role (specified in fam_vec or n_fam) as well as the lower and upper thresholds for all individuals in all families. Note that this tibble has the format required in estimate_liability. If either fam_vec or n_fam is used as the argument and if it is of the required format, if genetic_corrmat and full_corrmat are two numeric and symmetric matrices satisfying that all diagonal entries are one and that all off-diagonal entries are between -1 and 1, if the liability-scale heritabilities in h2_vec are numbers satisfying 0 \leq h^2_i for all i \in \{1,...,n_pheno\}, n_sim is a strictly positive number, and pop_prev is a positive numeric vector such that all entries are at most one, then the output will be a list containing the following lists. The first outer list, which is named after the first phenotype in phen_names, holds the tibble sim_obs, which holds the simulated liabilities, the disease status and the current age/age-of-onset for all family members in each of the n_sim families for the first phenotype. As the first outer list, the second outer list, which is named after the second phenotype in phen_names, holds the tibble sim_obs, which holds the simulated liabilities, the disease status and the current age/age-of-onset for all family members in each of the n_sim families for the second phenotype. There is a list containing sim_obs for each phenotype in phen_names. The last list entry, thresholds, holds the family identifier, the personal identifier, the role (specified in fam_vec or n_fam) as well as the lower and upper thresholds for all individuals in all families and all phenotypes. Note that this tibble has the format required in estimate_liability. Finally, note that if neither fam_vec nor n_fam are specified, the function returns the disease status, the current age/age-of-onset, the lower and upper thresholds, as well as the personal identifier for a single individual, namely the individual under consideration (called o). If both fam_vec and n_fam are defined, the user is asked to ' decide on which of the two vectors to use.

See Also

construct_covmat simulate_under_LTM_single simulate_under_LTM_multi

Examples

simulate_under_LTM()

genetic_corrmat <- matrix(0.4, 3, 3)
diag(genetic_corrmat) <- 1
full_corrmat <- matrix(0.6, 3, 3)
diag(full_corrmat) <- 1

simulate_under_LTM(fam_vec = NULL, n_fam = stats::setNames(c(1,1,1,2,2),
c("m","mgm","mgf","s","mhs")))

simulate_under_LTM(fam_vec = c("m","f","s1"), n_fam = NULL, add_ind = FALSE,
genetic_corrmat = genetic_corrmat, full_corrmat = full_corrmat, n_sim = 200)

simulate_under_LTM(fam_vec = c(), n_fam = NULL, add_ind = TRUE, h2 = 0.5,
n_sim = 200, pop_prev = 0.05)

Simulate under the liability threshold model (multiple phenotypes).

Description

simulate_under_LTM_multi simulates families and thresholds under the liability threshold model for a given family structure and multiple phenotypes. Please note that it is not possible to simulate different family structures.

Usage

simulate_under_LTM_multi(
  fam_vec = c("m", "f", "s1", "mgm", "mgf", "pgm", "pgf"),
  n_fam = NULL,
  add_ind = TRUE,
  genetic_corrmat = diag(3),
  full_corrmat = diag(3),
  h2_vec = rep(0.5, 3),
  phen_names = NULL,
  n_sim = 1000,
  pop_prev = rep(0.1, 3)
)

Arguments

Value

If either fam_vec or n_fam is used as the argument and if it is of the required format, if genetic_corrmat and full_corrmat are two numeric and symmetric matrices satisfying that all diagonal entries are one and that all off-diagonal entries are between -1 and 1, if the liability-scale heritabilities in h2_vec are numbers satisfying 0 \leq h^2_i for all i \in \{1,...,n_pheno\}, n_sim is a strictly positive number, and pop_prev is a positive numeric vector such that all entries are at most one, then the output will be a list containing lists for each phenotype. The first outer list, which is named after the first phenotype in phen_names, holds the tibble sim_obs, which holds the simulated liabilities, the disease status and the current age/age-of-onset for all family members in each of the n_sim families for the first phenotype. As the first outer list, the second outer list, which is named after the second phenotype in phen_names, holds the tibble sim_obs, which holds the simulated liabilities, the disease status and the current age/age-of-onset for all family members in each of the n_sim families for the second phenotype. There is a list containing sim_obs for each phenotype in phen_names. The last list entry, thresholds, holds the family identifier, the personal identifier, the role (specified in fam_vec or n_fam) as well as the lower and upper thresholds for all individuals in all families and all phenotypes. Note that this tibble has the format required in estimate_liability. Finally, note that if neither fam_vec nor n_fam are specified, the function returns the disease status, the current age/age-of-onset, the lower and upper thresholds, as well as the personal identifier for a single individual, namely the individual under consideration (called o). If both fam_vec and n_fam are defined, the user is asked to ' decide on which of the two vectors to use.

See Also

construct_covmat

Examples

simulate_under_LTM_multi()

genetic_corrmat <- matrix(0.4, 3, 3)
diag(genetic_corrmat) <- 1
full_corrmat <- matrix(0.6, 3, 3)
diag(full_corrmat) <- 1

simulate_under_LTM_multi(fam_vec = NULL, n_fam = stats::setNames(c(1,1,1,2,2),
c("m","mgm","mgf","s","mhs")))

simulate_under_LTM_multi(fam_vec = c("m","f","s1"), add_ind = FALSE,
genetic_corrmat = genetic_corrmat, full_corrmat = full_corrmat, n_sim = 100)

simulate_under_LTM_multi(fam_vec = c(), n_fam = NULL, add_ind = TRUE, n_sim = 150)

Simulate under the liability threshold model (single phenotype).

Description

simulate_under_LTM_single simulates families and thresholds under the liability threshold model for a given family structure and a single phenotype. Please note that it is not possible to simulate different family structures.

Usage

simulate_under_LTM_single(
  fam_vec = c("m", "f", "s1", "mgm", "mgf", "pgm", "pgf"),
  n_fam = NULL,
  add_ind = TRUE,
  h2 = 0.5,
  n_sim = 1000,
  pop_prev = 0.1
)

Arguments

Value

If either fam_vec or n_fam is used as the argument, if it is of the required format, if the liability-scale heritability h2 is a number satisfying 0 \leq h^2, n_sim is a strictly positive number, and pop_prev is a positive number that is at most one, then the output will be a list holding two tibbles. The first tibble, sim_obs, holds the simulated liabilities, the disease status and the current age/age-of-onset for all family members in each of the n_sim families. The second tibble, thresholds, holds the family identifier, the personal identifier, the role (specified in fam_vec or n_fam) as well as the lower and upper thresholds for all individuals in all families. Note that this tibble has the format required in estimate_liability. In addition, note that if neither fam_vec nor n_fam are specified, the function returns the disease status, the current age/age-of-onset, the lower and upper thresholds, as well as the personal identifier for a single individual, namely the individual under consideration (called o). If both fam_vec and n_fam are defined, the user is asked to ' decide on which of the two vectors to use.

See Also

construct_covmat, simulate_under_LTM_multi, simulate_under_LTM

Examples

simulate_under_LTM_single()
simulate_under_LTM_single(fam_vec = NULL, n_fam = stats::setNames(c(1,1,1,2),
c("m","mgm","mgf","mhs")))
simulate_under_LTM_single(fam_vec = c("m","f","s1"), n_fam = NULL, add_ind = FALSE,
h2 = 0.5, n_sim = 500, pop_prev = .05)
simulate_under_LTM_single(fam_vec = c(), n_fam = NULL, add_ind = TRUE, h2 = 0.5,
n_sim = 200, pop_prev = 0.05)

Title: Calculate the mean of the truncated normal distribution

Description

Title: Calculate the mean of the truncated normal distribution

Usage

tnorm_mean(mu = 0, sigma = 1, lower = -Inf, upper = Inf)

Arguments

Value

mean value of the truncated normal distribution

Examples

tnorm_mean()

Title: Calculates mean and variance of mixture of two truncated normal distributions

Description

Title: Calculates mean and variance of mixture of two truncated normal distributions

Usage

tnorm_mixture_conditional(mu, var, lower, upper, K_i, K_pop)

Arguments

Value

mean and variance of mixture distribution between two truncated normal distributions

Examples

tnorm_mixture_conditional(mu = 0, var = 1, lower = -Inf, upper = Inf, K_i = 0, K_pop = 0.01)
tnorm_mixture_conditional(mu = 0, var = 1, lower = -Inf, upper = 2, K_i = .01, K_pop = 0.05)

Title: Calculate the variance of the truncated normal distribution

Description

Title: Calculate the variance of the truncated normal distribution

Usage

tnorm_var(mu = 0, sigma = 1, lower = -Inf, upper = Inf)

Arguments

Value

mean value of the truncated normal distribution

Examples

tnorm_var()

CDF for truncated normal distribution.

Description

truncated_normal_cdf computes the cumulative density function for a truncated normal distribution.

Usage

truncated_normal_cdf(
  liability,
  lower = stats::qnorm(0.05, lower.tail = FALSE),
  upper = Inf
)

Arguments

Details

This function can be used to compute the value of the cumulative density function for a truncated normal distribution given an individual's true underlying liability.

Value

If liability is a number and the lower and upper cutoff points are numbers satisfying lower <= upper, then truncated_normal_cdf returns the probability that the liability will take on a value less than or equal to liability.

Examples

curve(sapply(liability, truncated_normal_cdf), from = qnorm(0.05, lower.tail = FALSE), to = 3.5,
 xname = "liability")