mJAM_get_PrM: Get Pr(Model) based on BF-type model probability
In USCbiostats/hJAM: Hierarchical Joint Analysis of Marginal Summary Statistics

mJAM_get_PrM

R Documentation

Get Pr(Model) based on BF-type model probability

Description

Also apply weighting to get robust estimates of yty

Usage

mJAM_get_PrM(
  GItGI,
  GIty,
  yty,
  yty_med,
  N_GWAS,
  C_id,
  prev_X_list = NULL,
  g = NULL,
  rare_SNPs = NULL,
  use_robust_var_est = FALSE
)

Arguments

`GItGI`	A list of transformed statistics from 'get_XtX()' for each study.
`GIty`	A list of transformed statistics from 'get_z()' for each study.
`yty`	A list of transformed statistics from 'get_yty()' for each study.
`yty_med`	A numeric vector of median yty across all SNPs within each study.
`N_GWAS`	A numeric vector of GWAS sample size for each study.
`C_id`	An ingeter vector of IDs for the SNPs to be tested.
`prev_X_list`	A numeric vector of the ID(s) of previously selected index SNP(s).
`g`	The pre-specified 'g' in 'g'-prior formulation.
`rare_SNPs`	A numeric vector of ID(s) for rare SNP(s) which we do not apply weighting. Instead, we use the individual estimate of yty for these SNPs for robustness.
`use_robust_var_est`	whether to use linear combination of median yty and individual yty.

Value

post_prob: Posterior Pr(Model) for each SNPs in 'C_id'.
R2_est: R2 estimates of every one-SNP model (one for each SNPs in 'C_id').
n_miss: An integer vector of how many studies have missing values for each SNP.