Description Usage Arguments Details Value Author(s) References See Also Examples
View source: R/fetchDiffCorrNetwork.R
take the input data from two conditions e.g. normalized gene expression data from normal and cancer cells,
and then compute the differential correlation network based on DiffCorr.
Correlation coefficients, pvalues and relation directions among entities in each condition are calculated using WGCNA functions cor
and corPvalueStudent
.
The correlation coefficients are continuous values between -1 (negative correlation) and 1 (positive correlation), with numbers close to 1 or -1, meaning very closely correlated.
Then correlation coefficients are test for differential correlations using Fisher's z-test based on DiffCorr.
1 | fetchDiffCorrNetwork(datX1, datX2, datY1, datY2, pDiff, method, returnAs)
|
datX1 |
data frame containing normalized, quantified omics data e.g. expression data, metabolite intensities of one condition. Columns correspond to entities e.g. genes, metabolites, and rows to samples. Require 'nodetype' at the first row to indicate the type of entities in each column. See below for details. |
datX2 |
data frame containing normalized, quantified omics data e.g. expression data, metabolite intensities of another condition.
Use the same format as |
datY1 |
data frame containing normalized, quantified omics data e.g. expression data, metabolite intensities of one condition.
Use the same format as |
datY2 |
data frame containing normalized, quantified omics data e.g. expression data, metabolite intensities of another condition.
Use the same format as |
pDiff |
numerical value to define the maximum value of pvalues (pvalDiff), to include edges in the output. |
method |
string to define which correlation is to be used. It can be one of "pearson","kendall","spearman", see |
returnAs |
string of output type. Specify the type of the returned network. It can be one of "tab","json","cytoscape". "cytoscape" is the format used in Cytoscape.js |
To calculate the differential correlation network, require the input data from two conditions; 1 and 2. The input data are matrices in which rows are samples and columns are entities.
For each condition, if datNormY is given, then the correlations between the columns of datNormX and the columns of datNormY are computed.
Otherwise if datNormY is not given, the correlations of the columns of datNormX are computed. Then correlation coefficients are test for significant correlation pairs.
If grinn functions will be used in further analyses, the column names of both datNormX and datNormY are suggested to use grinn ids.
convertToGrinnID
is provided for id conversion, see link{convertToGrinnID}
.
list of nodes and edges. The list is with the following componens: edges and nodes. Output includes correlation coefficients, pvalues and relation directions of each conditions, and the pvalues (pvalDiff) after testing. Return empty list if found nothing
Kwanjeera W kwanich@ucdavis.edu
Langfelder P. and Horvath S. (2008) WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics, 9:559
Dudoit S., Yang YH., Callow MJ. and Speed TP. (2002) Statistical methods for identifying differentially expressed genes in replicated cDNA microarray experiments, STATISTICA SINICA, 12:111
Langfelder P. and Horvath S. Tutorials for the WGCNA package http://labs.genetics.ucla.edu/horvath/CoexpressionNetwork/Rpackages/WGCNA/Tutorials/index.html
Fukushima A. (2013) DiffCorr: an R package to analyze and visualize differential correlations in biological networks. Gene, 10;518(1):209-14.
cor
, corPvalueStudent
, DiffCorr, http://js.cytoscape.org/
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 | # Compute a differential correlation network of metabolites
#dummyX1 <- rbind(nodetype=rep("metabolite"),mtcars[1:16,])
#colnames(dummyX1) <- c('G1.1','G27967','G371','G4.1',letters[1:7])
#rownames(dummyX1)[-1] <- paste0(rep("normal_"),1:16)
#dummyX2 <- rbind(nodetype=rep("metabolite"),mtcars[17:32,])
#colnames(dummyX2) <- c('G1.1','G27967','G371','G4.1',letters[1:7])
#rownames(dummyX2)[-1] <- paste0(rep("cancer_"),1:16)
#result <- fetchDiffCorrNetwork(datX1=dummyX1, datX2=dummyX2, datY1=NULL, datY2=NULL, pDiff=0.05, method="spearman", returnAs="tab")
#library(igraph)
#plot(graph.data.frame(result$edges[,1:2], directed=FALSE))
# Compute a differential correlation network of metabolites and proteins
#dummyX1 <- rbind(nodetype=rep("metabolite"),mtcars[1:16,1:5])
#colnames(dummyX1) <- c('G1.1','G27967','G371','G4.1','G16962')
#rownames(dummyX1)[-1] <- paste0(rep("normal_"),1:16)
#dummyX2 <- rbind(nodetype=rep("metabolite"),mtcars[17:32,1:5])
#colnames(dummyX2) <- c('G1.1','G27967','G371','G4.1','G16962')
#rownames(dummyX2)[-1] <- paste0(rep("cancer_"),1:16)
#dummyY1 <- rbind(nodetype=rep("protein"),mtcars[1:16,6:10])
#colnames(dummyY1) <- c('P28845','P08235','Q08AG9','P80365','P15538')
#rownames(dummyY1)[-1] <- paste0(rep("normal_"),1:16)
#dummyY2 <- rbind(nodetype=rep("protein"),mtcars[17:32,6:10])
#colnames(dummyY2) <- c('P28845','P08235','Q08AG9','P80365','P15538')
#rownames(dummyY2)[-1] <- paste0(rep("cancer_"),1:16)
#result <- fetchDiffCorrNetwork(datX1=dummyX1, datX2=dummyX2, datY1=dummyY1, datY2=dummyY2, pDiff=0.05, method="spearman", returnAs="tab")
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.