In liukf10/DDPNA: Disease-Drived Differential Proteins Co-Expression Network Analysis

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

DDPNA

'DDPNA' is a package for Disease-Drived Differential Proteins(DEP) and Proteome-Wide Co-expression Network Associated Analysis. The goal of 'DDPNA' is offered a better methods to analyze omic data. The package is designed for proteomic data, but it is also fit for expression data in RNA-seq and metabolome. It is associated DEP and co-expression network module, and constructed a Mod-DEP network to remove lower connectivity DEP. The lower connectivity DEP is hard to get the key function in PPI and is more likely a false postive protein. The Mod-DEP network can also get DEP related proteins which is more likely a false negative protein. It provides the essential statisic analysis included t.test, ANOVA analysis to extract differential proteins. The package also provide some module analysis included PCA analysis, two enrichment analysis, Planner maximally filtered graph extraction and hub analysis. The co-expression network should constructed by other package or software.('WGCNA' package or others)

Installation

You can install the developed version of DDPNA from github

install.packages("devtools")
devtools::install_github("liukf10/DDPNA")

Example

This is a basic example which shows you how to solve a common problem:

library(DDPNA)
##data extract and clean
data(Dforimpute)
#outlier sample remove and miss value impute
data <- Data_impute(Dforimpute, miss.value = 0, distmethod = "manhattan", plot = FALSE)
logD <- data$log2_value
rownames(logD) <- data$inf$ori.ID
## network construction
#net is constructed by WGCNA blockwiseModules function.
#the parameter:datExpr = t(logD), TOMType = "unsigned", deepSplit = 4, minModuleSize = 17, reassignThreshold = 0.05, mergeCutHeight = 0.07
data(net)
## module information extract
Module <- Module_inf(net, data$inf)
## differential protein analysis
oriData <- Dforimpute$LFQ
colnames(oriData) <- gsub("LFQ.intensity.","", colnames(oriData))
oriData <- oriData[,colnames(logD)]
rownames(oriData) <- Dforimpute$inf$ori.ID
group <- gsub("[0-9]+","", colnames(oriData))
up <- changedID(oriData, group, vs.set2 = "ad",vs.set1 = "ctl",
                rank = "foldchange",anova = FALSE, Padj = "none",cutoff = 1,
                datatype = "none",fctype = "up")
##module-DFP associated analysis
FCSenrich <- Module_Enrich(Module, up, datainf = rownames(oriData), coln="ori.ID")
FCSenrich <- FCSenrichplot(FCSenrich,plot=F)

pos <- which(net$colors == 4)
Mod4_PCA <- modpcomp(logD[pos,], net$colors[pos], plot = F, group = group)
Mod4 <- getmoduleHub(logD, Module, 4, coln = "ori.ID",adjustp = FALSE)
if (requireNamespace("MEGENA", quietly = TRUE)) {
  try(library(MEGENA), silent = TRUE)
  PMFG <- plot_subgraph(module = Mod4$degreeStat$gene,
                        hub = Mod4$hub, PFN = Mod4$PMFG,
                        node.default.color = "black",
                        gene.set = NULL, color.code = c("grey"),
                        show.legend = TRUE, label.hubs.only = TRUE,
                        hubLabel.col = "red", hubLabel.sizeProp = 0.5,
                        show.topn.hubs = 10, node.sizeProp = 13,
                        label.sizeProp = 13, label.scaleFactor = 10,
                        layout = "kamada.kawai")
  }

References

(1) Peter, L.; Steve H. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008, 9, 559.

(2) Won-Min Song, Bin Zhang. Multiscale Embedded Gene Co-expression Network Analysis PLOS Computational Biology. 2015, 11(11), e1004574

(3) Zhenfeng Wu, et al. NormExpression: an R package to normalize gene expression data using evaluated methods. BioRxiv 2018, JAN.

(4) Aravind Subramanian, et al. Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles PNAS 2005, 102(43), 15545-15550.

(5) Vince V. ggbiplot. 2018; https://github.com/vqv/ggbiplot.

(6) Michael C Oldham, et al. Network Methods for Describing Sample Relationships in Genomic Datasets: Application to Huntington's Disease; https://horvath.genetics.ucla.edu/html/CoexpressionNetwork/SampleNetwork/

liukf10/DDPNA documentation built on Sept. 29, 2021, 1:35 a.m.