QBLstrat: Quantitative Bayesian Lasso for Detecting Rare (or Common)...

In mxw010/LBL: Logistic and Quantitative Bayesian Lasso for identifying genetic association between common diseases and rare haplotypes

Description

QBLstrat is designed to identify rare (or common) haplotype effects associated with a quantitative trait. The software is able to adjust for population stratification using principal components (PCs). The input does not allow missing observations and subjects with missing data are removed. The function returns an object containing posterior inferences after the burn-in period.

Usage

QBLstrat(
  dat,
  numSNPs = 5,
  allelic = TRUE,
  baseline = "missing",
  interaction = F,
  cov = 1,
  pooling.tol = 0,
  zero.tol = "missing",
  a = 20,
  b = 20,
  start.beta = 0.01,
  lambda = 1,
  D = 0,
  seed = NULL,
  e = 0.1,
  burn.in = 20000,
  num.it = 50000,
  sigmas = 1,
  asig = 2,
  bsig = 1,
  CC = 1000,
  dumpConvergence = F,
  n.chains = 1,
  plot.interval = NULL
)

Arguments

dat	The data set should consist of n rows and 1+n.cov+2p (allelic format) or 1+n.cov+p (genotypic format) columns, where n is the number of individuals, n.cov is the number of covariates and PCs, and p is the number of SNPs. The first column is the trait, followed by covariates and PCs. If in allelic format, the other 2*p columns are allels with one column for each allele of the single-locus genotypes. If in genotypic format, the other p columns are genotypes with one column for each SNP. Missing data are not allowed and will be removed.
numSNPs	The number of SNPs which form the haplotypes.
allelic	TRUE if in allelic format and FALSE if in genotypic format; default is TRUE.
baseline	The name of the baseline haplotype (e.g., "h00000"); default is the one with the maximum frequency.
interaction	TRUE if consider the interaction between covariates and SNPs; only considers interaction with a maximum of the first 2 covariates; FALSE if else.
cov	The number of covariates and PCs.
pooling.tol	Haplotypes whose frequency is below pooling tolerance will be pooled together; default is 0
zero.tol	Tolerance for haplotype frequencies below which haplotypes are assumed not to exist; default is 1/(20*n) where n is the number of individuals.
a	First hyperparameter of the Gamma(a,b) prior for regression coefficients, β. The prior variance of β is 2/λ^2. The Gamma prior parameters a and b are formulated such that the mean and variance of the Gamma distribution are a/b and a/b^2. The default value of a is 20.
b	Second hyperparameter of the Gamma(a,b) distribution described above; default is 20.
start.beta	Starting value of all regression coefficients, β; default is 0.01.
lambda	Starting value of the λ parameter described above; default is 1.
D	Starting value of the D parameter, which is the within-population inbreeding coefficient; default is 0.
seed	Seed to be used for the MCMC in Bayesian Lasso; default is a random seed. If exact same results need to be reproduced, seed should be fixed to the same number.
e	A (small) number ε in the null hypothesis of no association, H_0: |β| ≤ ε. The default is 0.1. Changing e from the default of 0.1 may necessitate choosing a different threshold for Bayes Factor (one of the outputs) to infer association.
burn.in	Burn-in period of the MCMC sampling scheme; default is 20000.
num.it	Total number of MCMC iterations including burn-in; default is 50000.
sigmas	Starting value of the σ^2 parameter, which is the random error; default is 1.
asig	First hyperparameter of the Inverse-Gamma(asig, bsig) prior for the random error, σ^2. The prior mean and variance of σ^2 is bsig/(asig-1) and bsig^2/(asig-1)^2(asig-2), respectively. The default value of asig is 2.
bsig	Second hyperparameter of Inverse-Gamma(asig, bsig) described above; default is 1.
CC	The constant value set for the proposal distribution for updating haplotype frequencies, f. The proposal distribution of the MH algorithm is Dirichlet(a_1,...,a_p) with a_1+...+a_p=CC; default is 1000.
dumpConvergence	TRUE if check convergence; FALSE if else; default is FALSE.
n.chains	The number of chains for checking convergence; default if 1.
plot.interval	The interval between two posterior samples that are extracted to draw trace plots.

mxw010/LBL documentation built on Sept. 26, 2021, 3:44 a.m.