Reference

All functions

burren_imd_13_traits

Burren sparse basis files

burren_imd_13_traits

Burren sparse basis files

add_ref_annotations()

add_ref_annotations integrate GWAS summary data with support file

align_alleles()

align_alleles compute variance weighted euclidean distance between two PC loading vectors a and b

ambiguous()

ambiguous get ambiguous variants

burren_imd_13_traits

Burren sparse basis files

burren_imd_13_traits

Burren sparse basis files

comp()

comp get complement for a set of variants

compute_seb_proj_var()

analytically compute the variance of a projection given a standard error of beta for a given trait compute_seb_proj_var

compute_seb_proj_var_sparse()

analytically compute the variance of a projection given a standard error of beta for a given trait for sparse basis compute_seb_proj_var_sparse

compute_shrinkage_metrics()

This function computes various shrinkage metrics compute_shrinkage_metrics computes various shrinkage metrics

control_prior_shape()

This function computes shape parameters for beta distribution matching a binomial proportion B(n,f)/n control_prior_shape shape parameters for beta distribution matching a binomial proportion B(n,f)/n

cov_beta()

covariance matrix of betas for an ld block cov_beta

create_basis()

This function creates a basis create_basis creates a trait snp matrix that is suitable for basis creation and projection

create_ds_matrix()

This function creates a trait snp matrix create_ts_matrix creates a trait snp matrix that is suitable for basis creation and projection

e_lor()

This function computes expected log odds ratio for a set of shape parameters e_lor computes expected log odds ratio for a set of shape parameters

est_a1b1()

This function estimates shape parameters for a prior distribution of control allele frequencies est_a1b1 estimate shape parameters a1,b1

flip_allele()

align_allele given a data.table of pid,a1,a2 get a list of SNPs where alleles are flipped wrt to basis reference

fopt()

This function computes the optimal shape parameter a1 for a given set of constraints shape parameters estimate the P(OR > target.or). fopt estimate the P(OR > target.or) given shape parameters for case and control allele frequency

get_gwas_data()

get_gwas_data integrate GWAS summary data with support files

get_seb()

An internal helper function to get obtain standard error of beta estimates from a GWAS.DT data.frame object get_seb

logsum()

helper function to sum logs without loss of precision logsum sums logs without loss of precision

lor_constraint()

This function computes a probability for a given configuration of beta distribution shape parameters estimate the P(OR > target.or). lor_constraint estimate the P(OR > target.or) given shape parameters for case and control allele frequency

lor_f()

This function samples from the posterior distribution of a given control allele frequency, for different genotype configurations lor_f sample from log(or) posterior distribution given an allele frequency.

lyons_egpa

EGPA combined GWAS summary statistics

maf_se_estimate()

Compute minor allele frequency shrinkage maf_se_estimate computes a shrinkage metric for a given list of minor allele frequencies'

maf_se_estimate_sample_size()

Compute minor allele frequency shrinkage using sample size maf_se_estimate_sample_size computes component of standard error of beta due to minor allele frequency

mvs_perm()

Code to sample multivariate norm mvs_perm sample from a multivariate normal distribution

mvs_sigma()

Code to compute sigma - genotype covariance matrix mvs_sigma create a geneotype convariance matrix

opt_a1b1()

This function estimates shape parameters for a prior distribution of control allele frequencies opt_a1b1 estimate shape parameters a1,b1

p2z()

convert p value to a signed Z score p2z p value to a signed Z score

post_lor()

This function samples from the posterior distribution of log(OR) based on the observation of a case genotype post_lor samples from the posterior distribution of log(OR) given genotype.

project_basis()

This function projects an aligned trait onto the basis project_basis projects an external trait into basis space

project_sparse()

A streamlined function to project a trait onto a sparse basis project_sparse

burren_imd_13_traits

Burren sparse basis files

se_null()

This function computes standard error under the null se_null analytically compute standard error of \(\beta\) under \(\mathbb{E}(\beta) = 0\) \(\sqrt{\frac{1}{2fN_0} + \frac{1}{2N_{0}(1-f)} + \frac{1}{2fN_1} + \frac{1}{2N_{1}(1-f)} }\)

burren_imd_13_traits

Burren sparse basis files

simulate_beta()

simulate betas for an ld block simulate_beta use the multivariate normal to simulate realistic betas

simulate_study()

simulate betas for a study simulate_study use the multivariate normal to simulate realistic betas for a study

burren_imd_13_traits

Burren sparse basis files

vcf2snpmatrix()

convert a vcf file to snpMatrix object vcf2snpmatrix convert a vcf file to snpMatrix object

wakefield_null_pp()

compute reciprocal posterior probabilities using Wakefield's approximate Bayes Factors wakefield_null_pp computes posterior probabilities for a given SNP to be NOT be causal for a given SNP under the assumption of a single causal variant.

wakefield_pp()

compute posterior probabilities using Wakefield's approximate Bayes Factors wakefield_pp computes posterior probabilities for a given SNP to be causal for a given SNP under the assumption of a single causal variant.

ws_shrinkage()

This function computes an alternative to the Bayesian shrinkage method which can be too agressive. ws_shrinkage computes a shrinkage based on a weighted sum (ws) of posteriors for each disease this is then normalised by the total posterior for a given LD block

z2p()

convert z to p value p2z z to p value