ComputePValues: Compute p-values for affinity scores.

In chandlerzuo/atSNP: Affinity test for identifying regulatory SNPs

Description

This function computes the p-values for allele-specific affinity scores and between-allele affinity score changes using the importance sampling technique.

Usage

ComputePValues(
  motif.lib,
  snp.info,
  motif.scores,
  ncores = 1,
  testing.mc = FALSE,
  loglik.type = "max",
  figdir = NULL
)

Arguments

motif.lib	A list object with the output format of function LoadMotifLibrary.
snp.info	A list object with the output format of function LoadSNPData.
motif.scores	A data.frame object containing at least the following columns: motif The name of the motif. log_lik_ref The log-likelihood score for the reference allele. log_lik_snp The log-likelihood score for the SNP allele.
ncores	An integer for the number of parallel process. Default: 1.
testing.mc	If FALSE, Monte Carlo sample size is adaptively chosen between 2000 and 100K. If TRUE, the sample size is 100, which is used for testing purposes. Default: FALSE
loglik.type	A string of 'max', 'mean', 'median'.
figdir	A string for the path to print p-value plots for monitoring results. Default: NULL (no figure).

Value

A data.frame extending motif.scores by the following additional columns:

pval_ref	P-values for scores on the reference allele.
pval_snp	P-values for scores on the SNP allele.
pval_cond_ref	Conditional p-values for scores on the reference allele.
pval_cond_snp	Conditional p-values for scores on the SNP allele.
pval_diff	P-values for the difference in scores between the reference and the SNP alleles.
pval_rank	P-values for the log rank ratio between the reference and the SNP alleles.

Author(s)

Sunyoung Shin [e-mail], Chandler Zuo [e-mail]

Examples

data(example)
ComputePValues(motif_library, snpInfo, motif_scores$motif.scores, ncores = 2, testing.mc=TRUE)

chandlerzuo/atSNP documentation built on Oct. 31, 2020, 4:33 a.m.