non_partial_cor: Non-partial correlaton analysis

Description Usage Arguments Value Examples

Description

A method that integrates differential expression (DE) analysis and differential network (DN) analysis to select biomarker candidates for cancer studies. non_partial_cor is a one step function for user to perform the analysis, no pre-processing step required.

Usage

1
2
3
non_partial_cor(data = NULL, class_label = NULL, id = NULL,
  method = "pearson", p_val = NULL, permutation = 1000,
  permutation_thres = 0.025)

Arguments

data

This is a matrix of expression from all metabolites from all samples.

class_label

this is a binary array with 0 for group 1 and 1 for group 2.

id

This is an array of biomolecule ID to label.

method

This is a character string indicating which correlation coefficient is to be computed. The options are either "pearson" (the default) or "spearman".

p_val

This is optional, it is a data frame containing p-values for each metabolite/molecule.

permutation

This is a positive integer representing the desired number of permutations, default is 1000.

permutation_thres

This is a threshold for permutation. The defalut is 0.025 for each side to result in 95 percent confidience.

Value

A list containing a score data frame with "MetID", "P_value", "Node_Degree", "Activity_Score" and a differential network data frame with "Node1", "Node2", the binary link value and the weight link value.

Examples

1
2
non_partial_cor(data = Met_GU, class_label = Met_Group_GU, id = Met_name_GU,
                          method = "spearman", permutation_thres = 0.05, permutation = 1000)

kg737/INDEED_Patch documentation built on May 22, 2019, 6:32 p.m.