adding PerseusR support to MedianVar

MedianVar/MedianVarPerseus.R

@@ -0,0 +1,45 @@
+#devtools::install_github("cox-labs/PerseusR")
+library(PerseusR)
+args = commandArgs(trailingOnly=TRUE)
+if (length(args) != 2) {stop(paste("This script is for calculating the median of normalized variance from a Perseus matrix table.
+How to run:
+args[1] - Perseus matrix table of proteinGroup table
+args[2] - Row number of the annotation for sample groups
+args[3] - Row number of the annotation of Type which shows E, N, C and T
+args[4] - Row number of the first row of intensity values
+EX: Rscript MedianVar.R proteinGroup_Perseus.txt 4 1 5\n" ,sep="\n"), call.=FALSE)}
+cat("Running now...\n")
+inFile <- args[1]
+outFile <- args[2]
+#inFile<-"C:/Users/animeshs/GD/tools/MedianVar/proteinGroup_Perseus.txt"
+mdata <- read.perseus(inFile)
+Samples<-mdata@annotRows$Samples
+celltypes <- as.character(unique(Samples))
+Type<-mdata@annotRows$Experiments
+Channel<-mdata@annotRows$Channels
+Uniprot<-paste(mdata@annotCols$Protein.IDs)
+counts<- main(mdata)
+counts<-counts[complete.cases(counts), ]
+#countz<-counts
+colnames(counts)<-Samples
+vars<-c()
+a<-1
+for (cell in celltypes){
+  counts2 <- counts[,grepl(cell, names(counts))]
+  nor_vars_c <- c()
+  l<-1
+  for(j in seq(1, nrow(counts2))) {
+    nor_vars_c[l] <- var(unlist(counts2[j,]), na.rm=TRUE)/var(unlist(counts[j,]), na.rm=TRUE)
+    l<-l+1
+  }
+  vars[a]<-median(nor_vars_c,na.rm = T)
+  a<-a+1
+}
+#countz2<-counts2
+df <- data.frame(matrix(unlist(vars), nrow=length(vars), byrow=T))
+colnames(df)<-c("median of normalized variances")
+#print(df)
+outMdata <- matrixData(main=df,annotCols=as.data.frame(matrix(unlist(celltypes), nrow=length(vars), byrow=T)))
+#outFile<-"C:/Users/animeshs/GD/tools/MedianVar/proteinGroup_Perseus.txt.out.txt"
+write.perseus(outMdata,annotCols=celltypes, outFile)
+cat("Done...\n")

README.md

@@ -1,4 +1,5 @@
 # tools
+
 ## MedianVar.R
 This script is for calculating the median of normalized variance from a Perseus matrix table.
 
@@ -8,3 +9,24 @@ args[2] - Row number of the annotation for sample groups<br/>
 args[3] - Row number of the annotation of Type which shows E, N, C and T<br/>
 args[4] - Row number of the first row of intensity values<br/>
 EX: Rscript MedianVar.R proteinGroup_Perseus.txt 4 1 5<br/>
+
+## MedianVarPerseus.R
+This script is for calculating same the above using [External:Matrix=>R](https://github.com/cox-labs/PluginInterop) [PerseusR](https://github.com/cox-labs/PerseusR) to call directly from [Perseus](https://maxquant.net/download_asset/perseus/1.6.15.0) after installing R: 
+```
+devtools::install_github("cox-labs/PerseusR")
+```
+
+Test:
+
+```
+File Generic-matrix-upload proteinGroup_Perseus.txt into Perseus 
+Invoke External:Matrix=>R and provide script&R location like following respectively:
+script: <full-path-to>MedianVarPerseus.R
+R: <full-path-to>Rscript.exe
+```
+As shown in session1.sps.
+
+PS: on windows if "Error in utils::download.file(url, path, method = method, quiet = quiet,  :cannot open URL 'https://api.github.com/repos/cox-labs/PerseusR/tarball/HEAD'" run "options(download.file.method = "wininet")" before installing:
+
+## aovPerseusR\proteinGroupsANOVA.r
+R script callable via https://github.com/cox-labs/PerseusR as above that can reproduce the ANOVA p&q-values(using Benjamini/Hochberg) from [Multiple-sample tests in Perseus](http://coxdocs.org/doku.php?id=perseus:user:activities:MatrixProcessing:Tests:MultipleSampleTestProcessing): It also provides the mean-difference with posthoc/TukeyHSD corrected p-values for each group comparison as shown in the attached Perseus session. Need to be mindful that NAs are being imputed as 0 so better to take care of NAs in Perseus beforehand! 

aovPerseusR/LICENSE.md

@@ -0,0 +1,20 @@
+Copyright (c) 2021- @animesh .
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

aovPerseusR/README.md

@@ -0,0 +1,41 @@
+## proteinGroupsANOVA.r
+R script callable via https://github.com/cox-labs/PerseusR that can reproduce the ANOVA p&q-values(using Benjamini/Hochberg) from [Multiple-sample tests in Perseus](http://coxdocs.org/doku.php?id=perseus:user:activities:MatrixProcessing:Tests:MultipleSampleTestProcessing): It also provides the mean-difference with posthoc/TukeyHSD corrected p-values for each group comparison as shown in the attached Perseus session. Need to be mindful that NAs are being imputed as 0 so better to take care of NAs in Perseus beforehand! 
+
+## pre-requisites
+[R](https://cloud.r-project.org/), [devtools](https://www.r-project.org/nosvn/pandoc/devtools.html) and [PerseusR](https://github.com/cox-labs/PerseusR) , note on windows if 
+`"Error in utils::download.file(url, path, method = method, quiet = quiet,  :cannot open URL 'https://api.github.com/repos/cox-labs/PerseusR/tarball/HEAD'"
+`use
+`options(download.file.method = "wininet")`
+followed by install
+`devtools::install_github("cox-labs/PerseusR")` within R session
+
+## usage
+Download [Perseus](https://maxquant.net/download_asset/perseus/latest) and invoke [External:Matrix=>R](https://github.com/cox-labs/PluginInterop) with following parameters:
+script: <full path to>\proteinGroupsANOVA.r
+R: <full path to>\Rscript.exe
+
+## test
+```
+aovt<-anova(lm(c(-0.267558, -1.34843, -0.736034, -0.950788, -0.631172, -1.35648, -0.338978, 0.179467, -1.38368, -0.374963, -0.0118891, 0.583722, 0.646115, 2.28853, -0.117149)~c(rep("g1",5),rep("g2",5),rep("g3",5))))
+summary.lm(aovt)
+aovt<-aov(c(-0.267558, -1.34843, -0.736034, -0.950788, -0.631172, -1.35648, -0.338978, 0.179467, -1.38368, -0.374963, -0.0118891, 0.583722, 0.646115, 2.28853, -0.117149)~c(rep("g1",5),rep("g2",5),rep("g3",5))) #0.0134
+summary(aovt)[[1]][["Pr(>F)"]][[1]]
+postHoc<-TukeyHSD(aovt,ordered = TRUE)
+names(postHoc)<-"compare"
+postHoc<-data.frame(postHoc$compare)
+diff<-t(postHoc["diff"])
+plot(TukeyHSD(aovt))
+padj<-sapply(postHoc["p.adj"],as.numeric)
+-log10(padj) 
+        p.adj
+[1,] 0.019911
+[2,] 1.720245
+[3,] 1.495404
+```
+
+comparison to Perseus session session.sps file:
+```
+Group1	Group2	Group3	C: ANOVA Significant	N: ANOVA p value	N: ANOVA q-value	T: Name	T: Significant pairs
+-1.72025	-1.49541	1.72025		0.0134469	0.644	Row 12	Group3_Group1;Group3_Group2
+```
+but note that first values/-1.72025 in Perseus is -(thirdvalue)/1.72025 though Group3 could very well be -Group2 i.e. log10(g3-g2) =>  -1.4954 which is log10(0.0319592) from R. Probably it was decided to base G3 from g3-g1 and not g3-g3 because of the fact that 0.0190439 < 0.0319592 ? Unfortunately, the [Source code: not public](http://coxdocs.org/doku.php?id=perseus:user:activities:MatrixProcessing:Tests:MultipleSampleTestProcessing), hope it soon is https://github.com/cox-labs @JurgenCox