calculate.rho.signed: correlation calculation

View source: R/CorrelationScreen.R

calculate.rho.signedR Documentation

correlation calculation

Description

correlation analysis with FDR calculation

Usage

calculate.rho.signed(datExpr,n.perm = 10,FDR.cutoff = 0.05,estimator = "pearson",
                               use.obs = "na.or.complete",
                               direction = "absolute",
                               rho.thresh = NULL,sort.el = TRUE)

Arguments

datExpr

gene expression data matrix

n.perm

Number of permutations to perform. If doPerm = NULL, calculates BH FDR p-values instead of permutation based FDR.

FDR.cutoff

FDR threshold to output final results of significant correlations.

estimator

Type of correlation coefficient to calculate, and gets passed to 'method' in 'cor()' function. Either "pearson" or "spearman".

direction

If direction = "absolute", absolute valued correlations are considered. If direction = "positive" or "negative", signed correlation is considered, and significance is calculated for positive/negative values.

rho.thresh

vector of correlation cutoff threshold to calculate significance. If NULL, threshold values will be generated by increment of 0.01 between -1 and 1 (if direction = "positive" or "negative"), or 0 and 1 (if direction = "absolute").

sort.el

logical value to determine whether correlation list should be sorted

use.obs

an optional character string giving a method for computing covariances in the presence of missing values. This must be (an abbreviation of) one of the strings "everything", "all.obs", "complete.obs", "na.or.complete", or "pairwise.complete.obs".

Details

direction = "absolute" can serve as two-tailed significance test, where as direction = "positive"/"negative" is one-tailed test.

Value

output is a list. output$signif.ijw is three column edgelist data.frame, third column being the weight. output$FDR is the permutation FDR table across the threshold values in rho.thresh.

Author(s)

Won-Min Song

Examples


data(Sample_Expression)
rho.out = calculate.rho.signed(datExpr,n.perm = 10,FDR.cutoff = 0.05,estimator = "pearson",
                               use.obs = "na.or.complete",
                               direction = "absolute",
                               rho.thresh = NULL,sort.el = TRUE)


songw01/MEGENA documentation built on Jan. 19, 2024, 6:51 p.m.