differential_count_models: rgedgeRpaired

comparison rgedgeRpaired_nocamera.xml @ 77:4a2e7a9725b2 draft

Uploaded

author	fubar
date	Tue, 25 Feb 2014 23:54:59 -0500
parents	151bf55e018a
children	340d5460f3ff

comparison

equal deleted inserted replaced

-:2a377f98ab76
+:4a2e7a9725b2
-<tool id="rgDifferentialCount" name="Differential_Count" version="0.31">
+<tool id="rgDifferentialCount" name="Differential_Count" version="0.22">
 <description>models using BioConductor packages</description>
 <requirements>
 <requirement type="package" version="2.14">biocbasics</requirement>
 <requirement type="package" version="3.0.2">r302</requirement>
 <requirement type="package" version="1.3.18">graphicsmagick</requirement>
-<requirement type="package" version="9.10">ghostscript</requirement>
+<requirement type="package" version="9.07">ghostscript</requirement>
 </requirements>
 <command interpreter="python">
 rgToolFactory.py --script_path "$runme" --interpreter "Rscript" --tool_name "DifferentialCounts"
 --output_dir "$html_file.files_path" --output_html "$html_file" --make_HTML "yes"
 help="edgeR uses a negative binomial model and seems to be powerful, even with few replicates">
 <option value="F">Do not run edgeR</option>
 <option value="T" selected="true">Run edgeR</option>
 </param>
 <when value="T">
-<param name="edgeR_priordf" type="integer" value="20" size="3"
+<param name="edgeR_priordf" type="integer" value="10" size="3"
-label="prior.df for tagwise dispersion - lower value = more emphasis on each tag's variance. Replaces prior.n  and prior.df = prior.n * residual.df"
+label="prior.df for tagwise dispersion - larger value = more squeezing of tag dispersions to common dispersion. Replaces prior.n  and prior.df = prior.n * residual.df"
-help="0 = Use edgeR default. Use a small value to 'smooth' small samples. See edgeR docs and note below"/>
+help="10 = edgeR default. Use a larger value to 'smooth' small samples. See edgeR docs and note below"/>
+<param name="edgeR_robust" type="select" value="20" size="3"
+label="Use robust dispersion method"
+help="Use ordinary, anscombe or deviance robust deviance estimates">
+<option value="ordinary" selected="true">Use ordinary deviance estimates</option>
+<option value="deviance">Use robust deviance estimates</option>
+<option value="anscombe">use Anscombe robust deviance estimates</option>
+</param>
 </when>
 <when value="F"></when>
 </conditional>
 <conditional name="DESeq2">
 <param name="doDESeq2" type="select"
 <param name='doedgeR' value='T' />
 <param name='doVoom' value='T' />
 <param name='doDESeq2' value='T' />
 <param name='fdrtype' value='fdr' />
 <param name='edgeR_priordf' value="8" />
+<param name='edgeR_robust' value="ordinary" />
 <param name='fdrthresh' value="0.05" />
 <param name='control_name' value='heart' />
 <param name='subjectids' value='' />
 <param name='Control_cols' value='3,4,5,9' />
 <param name='Treat_cols' value='2,6,7,8' />
 <configfiles>
 <configfile name="runme">
 <![CDATA[
 #
 # edgeR.Rscript
+# updated feb 2014 adding outlier-robust deviance estimate options by ross for R 3.0.2/bioc 2.13
 # updated npv 2011 for R 2.14.0 and edgeR 2.4.0 by ross
 # Performs DGE on a count table containing n replicates of two conditions
 #
 # Parameters
 #
 edgeIt = function (Count_Matrix=c(),group=c(),out_edgeR=F,out_VOOM=F,out_DESeq2=F,fdrtype='fdr',priordf=5,
 fdrthresh=0.05,outputdir='.', myTitle='Differential Counts',libSize=c(),useNDF=F,
 filterquantile=0.2, subjects=c(),mydesign=NULL,
 doDESeq2=T,doVoom=T,doCamera=T,doedgeR=T,org='hg19',
 histgmt="", bigmt="/data/genomes/gsea/3.1/Abetterchoice_nocgp_c2_c3_c5_symbols_all.gmt",
-doCook=F,DESeq_fitType="parameteric")
+doCook=F,DESeq_fitType="parameteric",robustmeth='ordinary')
 {
 # Error handling
 if (length(unique(group))!=2){
 print("Number of conditions identified in experiment does not equal 2")
 q()
 if (doedgeR) {
 sink('edgeR.log')
 #### Setup DGEList object
 DGEList = DGEList(counts=workCM, group = group)
 DGEList = calcNormFactors(DGEList)
+if (robust_meth == 'ordinary') {
 DGEList = estimateGLMCommonDisp(DGEList,mydesign)
-comdisp = DGEList\$common.dispersion
+DGEList = estimateGLMTrendedDisp(DGEList,mydesign)
-DGEList = estimateGLMTrendedDisp(DGEList,mydesign)
+DGEList = estimateGLMTagwiseDisp(DGEList,mydesign,prior.df = edgeR_priordf)
-if (edgeR_priordf > 0) {
-print.noquote(paste("prior.df =",edgeR_priordf))
+comdisp = DGEList\$common.dispersion
-DGEList = estimateGLMTagwiseDisp(DGEList,mydesign,prior.df = edgeR_priordf)
+estpriorn = getPriorN(DGEList)
-} else {
+print(paste("Common Dispersion =",comdisp,"CV = ",sqrt(comdisp),"getPriorN = ",estpriorn),quote=F)
-DGEList = estimateGLMTagwiseDisp(DGEList,mydesign)
+} else {
+DGEList = estimateGLMRobustDisp(DGEList,design=mydesign, prior.df = edgeR_priordf, maxit = 6, residual.type = robust_meth)
+}
 }
 DGLM = glmFit(DGEList,design=mydesign)
 DE = glmLRT(DGLM,coef=ncol(DGLM\$design)) # always last one - subject is first if needed
 efflib = DGEList\$samples\$lib.size*DGEList\$samples\$norm.factors
 normData = (1e+06*DGEList\$counts/efflib)
 pdf("edgeR_GoodnessofFit.pdf")
 qq = qqnorm(z, panel.first=grid(), main="tagwise dispersion")
 abline(0,1,lwd=3)
 points(qq\$x[goodness\$outlier],qq\$y[goodness\$outlier], pch=16, col="maroon")
 dev.off()
-estpriorn = getPriorN(DGEList)
-print(paste("Common Dispersion =",comdisp,"CV = ",sqrt(comdisp),"getPriorN = ",estpriorn),quote=F)
 efflib = DGEList\$samples\$lib.size*DGEList\$samples\$norm.factors
 normData = (1e+06*DGEList\$counts/efflib)
 uniqueg = unique(group)
 #### Plot MDS
 sample_colors =  match(group,levels(group))
 #DESeq2 = DESeq(deSEQds,fitType='local',pAdjustMethod=fdrtype)
 #rDESeq = results(DESeq2)
 #newCountDataSet(workCM, group)
 deSeqDatsizefac = estimateSizeFactors(deSEQds)
 deSeqDatdisp = estimateDispersions(deSeqDatsizefac,fitType=DESeq_fitType)
-resDESeq = nbinomWaldTest(deSeqDatdisp)
+resDESeq = nbinomWaldTest(deSeqDatdisp, pAdjustMethod=fdrtype)
 rDESeq = as.data.frame(results(resDESeq))
 rDESeq = cbind(Contig=rownames(workCM),rDESeq,NReads=cmrowsums,URL=contigurls)
 srDESeq = rDESeq[order(rDESeq\$pvalue),]
 qqPlot(descr=paste(myTitle,'DESeq2 adj p qq plot'),pvector=rDESeq\$padj,outpdf='DESeq2_qqplot.pdf')
 cat("# DESeq top 50\n")
 if (doVoom == T) {
 sink('VOOM.log')
 if (doedgeR == F) {
 #### Setup DGEList object
 DGEList = DGEList(counts=workCM, group = group)
-DGEList = calcNormFactors(DGEList)
 DGEList = estimateGLMCommonDisp(DGEList,mydesign)
 DGEList = estimateGLMTrendedDisp(DGEList,mydesign)
 DGEList = estimateGLMTagwiseDisp(DGEList,mydesign)
-DGEList = estimateGLMTagwiseDisp(DGEList,mydesign)
-norm.factor = DGEList\$samples\$norm.factors
 }
+norm.factor = calcNormFactors(DGEList)
 pdf("VOOM_mean_variance_plot.pdf")
 dat.voomed = voom(DGEList, mydesign, plot = TRUE, lib.size = colSums(workCM) * norm.factor)
 dev.off()
 # Use limma to fit data
 fit = lmFit(dat.voomed, mydesign)
 history_gmt = ""
 history_gmt_name = ""
 out_edgeR = F
 out_DESeq2 = F
 out_VOOM = "$out_VOOM"
+edgeR_robust_meth = $edgeR_robust # control robust deviance options
 doDESeq2 = $DESeq2.doDESeq2 # make these T or F
 doVoom = $doVoom
 doCamera = F
 doedgeR = $edgeR.doedgeR
-edgeR_priordf = 0
+edgeR_priordf = 10
 #if $doVoom == "T":
 out_VOOM = "$out_VOOM"
 #end if
 colnames(Count_Matrix) = paste(group,colnames(Count_Matrix),sep="_")                   #Relable columns
 results = edgeIt(Count_Matrix=Count_Matrix,group=group, out_edgeR=out_edgeR, out_VOOM=out_VOOM, out_DESeq2=out_DESeq2,
 fdrtype='BH',mydesign=NULL,priordf=edgeR_priordf,fdrthresh=fdrthresh,outputdir='.',
 myTitle=myTitle,useNDF=F,libSize=c(),filterquantile=fQ,subjects=subjects,
 doDESeq2=doDESeq2,doVoom=doVoom,doCamera=doCamera,doedgeR=doedgeR,org=org,
-histgmt=history_gmt,bigmt=builtin_gmt,DESeq_fitType=DESeq_fitType)
+histgmt=history_gmt,bigmt=builtin_gmt,DESeq_fitType=DESeq_fitType,robustmeth=edgeR_robust_meth)
 sessionInfo()
 ]]>
 </configfile>
 </configfiles>
 <help>

Mercurial > repos > fubar > differential_count_models

comparison rgedgeRpaired_nocamera.xml @ 77:4a2e7a9725b2 draft