Combines SNP-level dAD results to gene-level.

AdvancedPvalcomb relies on the output of maelstRom's `dAD_analysis` function, applied on SNP-level data, and combines them to SNP-level p-values. Written to be run in parallel through the parallel's package `parLapply` function.

AdvancedPvalcomb(DATAS, SE, nperm = 10000)

Arguments

DATAS

List. A list containing four elements: A vector of (unique) gene names whose SNP-level results are being combined (1). A list of subsets of the `dAD_analysis` output dataframe containing, for every gene (see first list-element), the SNP-level results for SNPs assigned to said gene (2). A list of lists of dataframes containing, for every gene, a list of per-SNP dataframes containing data for control samples (columns "Sample", "Outlier", "ref_count", and "var_count"; respectively sample-IDs, an Outlier indicator (0 if no outlier, 1 if so), reference-, and variant allele counts) (3). A list of lists of dataframes containing, for every gene, a list of per-SNP dataframes containing data for case samples (columns "Sample", "Outlier", "ref_count", and "var_count"; respectively sample-IDs, an Outlier indicator (0 if no outlier, 1 if so), reference-, and variant allele counts) (4).

SE

Number. Sequencing error.

nperm

Integer. Number of iterations used for correlation-corrected Lancaster p-value combination, which estimates the dependence between samples through a (large) number of sample permutations (default 10000).

Value

A list containing the following components:

FisherPVAL_vec

Vector containing, for all genes, combined p-values using the Fisher combination method.

LancasterPVAL_vec

Vector containing, for all genes, combined p-values using the Lancaster combination method.

CorLancasterPVAL_vec

Vector containing, for all genes, combined p-values using the dependence(correlation)-corrected Lancaster combination method.

CauchyPVAL_vec

Vector containing, for all genes, combined p-values using the Cauchy combination method.

CLPV_OffDiagSum_vec

Vector containing, for all genes, the sum of between-sample correlations as estimated during the dependence corrected Lancaster combination method; a sort of measure for the amount of between-sample dependence present amongst SNPs.