FastWGCNA: Standard pipeline for WGCNA

In shijianasdf/BasicBioinformaticsAnalysisFromZhongShan: Launch Usual Clinical tool for Kind Yield

Description

FastWGCNA give a fast and standard pipeline to do Weighted Correlation Network Analysis(WGCNA) for specified expression matrix and design object

Usage

FastWGCNA(expr.matrix, design, log.convert = T,
  contrast.col = "N.status", contrast.control = "N0", check = F,
  mad.portion = 0.75, mad.min = 0.01, verbose = c(goodSamplesGenes =
  3, pickSoftThreshold = 5, blockwiseModules = 3), maxBlockSize = 50000,
  minModuleSize = 30, cutoff.pval = 0.05, corType = c("pearson",
  "bicor")[1], hub_cutoffSigGM = 0.2, hub_MM = 0.8,
  hub_WeightedQ = 0.01, parallel = F, report = T, two.step = F,
  save.path = "WGCNA", names = "love")

Arguments

expr.matrix	expression matrix
design	a design object
log.convert	whether do log2 scale for expression matrix
contrast.col	the colname of contrast in design object like "N.status"
contrast.control	the contrast control like "N0"
check	whether to check critical objecta in save-file space:"dataExpr.rda","Net of WGCNA.rda","soft-thresholding list.rda".Set check=T help continue job based on the previous efforts
mad.portion	If mad.portion <= 1, it represent the portion of data you want base on median absolute deviation(mad) filter.If mad.portion >1(for example 3000),it means the first mad.portion mad genes would be selected to use.
mad.min	the lower cut-off of mad filter
verbose	a named vector of verboses in multiple scenes
maxBlockSize	set as large as possible according to the internal storage of your computer
minModuleSize	the min count of module.Default is 30
cutoff.pval	cut-off of the p value in significant module-phenotype filter
corType	One of "pearson" and "bicor".Default is "pearson"
hub_cutoffSigGM	the cut-off of significant genes-Modules in hub genes exploration.Default is 0.2
hub_MM	the cut-off of Module Memberships in hub genes exploration.Default is 0.8
hub_WeightedQ	the cut-off of weighted q value in hub genes exploration.Default is 0.01
parallel	whether use enableWGCNAThreads() to accelarate cor or bicor functions
report	whether to do a plot report
two.step	whether use two step to complete calculation.If the nGenes is large,"two.step = T" is recommaned.
save.path	the space of the save file.Default is "WGCNA"
names	part of saved files name

Note

1.Please choose a high-performance computer for better running. /n 2.properly set "maxBlockSize" and "mad.portion" parameters to avoid the different block produced.

Author(s)

Weibin Huang<[e-mail]>

Examples

## This is a simulative process and NOT RUN

## data preparation
load1(c("fpkm","design.model2"))
set.seed(2018);select <- sample(1:nrow(data.fpkm),2000)
expr.matrix = data.fpkm[select,]
design = design.model2

## Quick Start
wgcna <- FastWGCNA(expr.matrix,
                  design,
                  contrast.col = "N.status",
                  contrast.control = "N0",
                  check = T,
                  parallel = F,
                  save.path = "WGCNA",
                  names = "test1");;mymusic(1)

shijianasdf/BasicBioinformaticsAnalysisFromZhongShan documentation built on Jan. 3, 2020, 10:08 p.m.