bryan/hartman-server
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity

Generalize REMc analysis

Browse Source
This commit is contained in:
Bryan Roessler
2024-07-28 05:19:50 -04:00
parent c8eba4efd4
commit 8de5005055
8 changed files with 179 additions and 11994 deletions
Download Patch File Download Diff File Expand all files Collapse all files

115
workflow/apps/r/joinInteractExps.R
View File

@@ -1,3 +1,4 @@
#!/usr/bin/env Rscript
# JoinInteractExps.R
library(plyr)
@@ -46,40 +47,40 @@ print(length(inputFiles))
# Join the two files at a time as a function of how many inputFile, list the larger file first ? in this example X2 has the larger number of genes.
# If X1 has a larger number of genes, switch the order of X1 and X2
if(length(inputFiles)==2) {
X1 <- read.csv(file=inputFiles[1],stringsAsFactors = FALSE)
X2 <- read.csv(file=inputFiles[2],stringsAsFactors = FALSE)
X1 <- read.csv(file=inputFiles[1],stringsAsFactors=FALSE)
X2 <- read.csv(file=inputFiles[2],stringsAsFactors=FALSE)
X <- join(X1,X2,by="OrfRep")
OBH=X[,order(colnames(X))] #OrderByHeader
headSel= select(OBH, contains('OrfRep'), matches('Gene'), contains('Z_lm_K'), contains('Z_Shift_K'),contains('Z_lm_L'), contains('Z_Shift_L'))
headSel= select(headSel, -'Gene.1') #remove 'Gene.1 column
headSel2 = select(OBH, contains('OrfRep'), matches('Gene')) #Frame for interleaving Z_lm with Shift colums
headSel2 = select(headSel2, -'Gene.1') #remove 'Gene.1 column #Frame for interleaving Z_lm with Shift colums
headSel=select(OBH, contains('OrfRep'), matches('Gene'), contains('Z_lm_K'), contains('Z_Shift_K'),contains('Z_lm_L'), contains('Z_Shift_L'))
headSel=select(headSel, -'Gene.1') #remove 'Gene.1 column
headSel2=select(OBH, contains('OrfRep'), matches('Gene')) #Frame for interleaving Z_lm with Shift colums
headSel2=select(headSel2, -'Gene.1') #remove 'Gene.1 column #Frame for interleaving Z_lm with Shift colums
}else if(length(inputFiles)==3){
X1 <- read.csv(file=inputFiles[1],stringsAsFactors = FALSE) #exp1File,stringsAsFactors = FALSE)
X2 <- read.csv(file=inputFiles[2],stringsAsFactors = FALSE) #exp2File,stringsAsFactors = FALSE)
X3 <- read.csv(file=inputFiles[3],stringsAsFactors = FALSE) #exp3File,stringsAsFactors = FALSE)
X1 <- read.csv(file=inputFiles[1],stringsAsFactors=FALSE) #exp1File,stringsAsFactors=FALSE)
X2 <- read.csv(file=inputFiles[2],stringsAsFactors=FALSE) #exp2File,stringsAsFactors=FALSE)
X3 <- read.csv(file=inputFiles[3],stringsAsFactors=FALSE) #exp3File,stringsAsFactors=FALSE)
X <- join(X1,X2,by="OrfRep")
X <- join(X,X3,by="OrfRep")
OBH=X[,order(colnames(X))] #OrderByHeader
headSel= select(OBH, contains('OrfRep'), matches('Gene'), contains('Z_lm_K'), contains('Z_Shift_K'),contains('Z_lm_L'), contains('Z_Shift_L'))
headSel= select(headSel, -'Gene.1',-'Gene.2')
headSel2 = select(OBH, contains('OrfRep'), matches('Gene'))
headSel2 = select(headSel2, -'Gene.1',-'Gene.2')
headSel=select(OBH, contains('OrfRep'), matches('Gene'), contains('Z_lm_K'), contains('Z_Shift_K'),contains('Z_lm_L'), contains('Z_Shift_L'))
headSel=select(headSel, -'Gene.1',-'Gene.2')
headSel2=select(OBH, contains('OrfRep'), matches('Gene'))
headSel2=select(headSel2, -'Gene.1',-'Gene.2')
}else if(length(inputFiles)==4){
X1 <- read.csv(file=inputFiles[1],stringsAsFactors = FALSE) #exp1File,stringsAsFactors = FALSE)
X2 <- read.csv(file=inputFiles[2],stringsAsFactors = FALSE) #exp2File,stringsAsFactors = FALSE)
X3 <- read.csv(file=inputFiles[3],stringsAsFactors = FALSE) #exp3File,stringsAsFactors = FALSE)
X4 <- read.csv(file=inputFiles[4],stringsAsFactors = FALSE) #exp4File,stringsAsFactors = FALSE)
X1 <- read.csv(file=inputFiles[1],stringsAsFactors=FALSE) #exp1File,stringsAsFactors=FALSE)
X2 <- read.csv(file=inputFiles[2],stringsAsFactors=FALSE) #exp2File,stringsAsFactors=FALSE)
X3 <- read.csv(file=inputFiles[3],stringsAsFactors=FALSE) #exp3File,stringsAsFactors=FALSE)
X4 <- read.csv(file=inputFiles[4],stringsAsFactors=FALSE) #exp4File,stringsAsFactors=FALSE)
X <- join(X1,X2,by="OrfRep")
X <- join(X,X3,by="OrfRep")
X <- join(X,X4,by="OrfRep")
OBH=X[,order(colnames(X))] #OrderByHeader
headSel= select(OBH, contains('OrfRep'), matches('Gene'), contains('Z_lm_K'), contains('Z_Shift_K'),contains('Z_lm_L'), contains('Z_Shift_L'))
headSel= select(headSel, -'Gene.1',-'Gene.2',-'Gene.3')
headSel2 = select(OBH, contains('OrfRep'), matches('Gene'))
headSel2 = select(headSel2, -'Gene.1',-'Gene.2',-'Gene.3')
headSel=select(OBH, contains('OrfRep'), matches('Gene'), contains('Z_lm_K'), contains('Z_Shift_K'),contains('Z_lm_L'), contains('Z_Shift_L'))
headSel=select(headSel, -'Gene.1',-'Gene.2',-'Gene.3')
headSel2=select(OBH, contains('OrfRep'), matches('Gene'))
headSel2=select(headSel2, -'Gene.1',-'Gene.2',-'Gene.3')
}
#headSel$contains('Z_Shift') %>% replace_na(0.001)
@@ -88,10 +89,10 @@ i=0
for(i in 1:length(headers)){
if(grepl("Shift",headers[i])) {
headSel[headers[i]][is.na(headSel[headers[i]])] = 0.001
headSel[headers[i]][is.na(headSel[headers[i]])]=0.001
}
if(grepl("Z_lm_",headers[i])) {
headSel[headers[i]][is.na(headSel[headers[i]])] = 0.0001
headSel[headers[i]][is.na(headSel[headers[i]])]=0.0001
}
}
@@ -101,7 +102,7 @@ REMcRdy=select(headSel, contains('OrfRep'), matches('Gene'), contains('Z_lm_'))
shiftOnly=select(headSel, contains('OrfRep'), matches('Gene'), contains('Z_Shift'))
# Code to replace the numeric (.1 .2 .3) headers with experiment names from StudyInfo.txt
Labels <- read.csv(file= "../Code/StudyInfo.csv",stringsAsFactors = FALSE,sep= ",")
Labels <- read.csv(file="../Code/StudyInfo.csv",stringsAsFactors=FALSE,sep=",")
# Using Text search grepl to relabel headers
REMcRdyHdr=colnames(REMcRdy)
@@ -152,25 +153,25 @@ for(i in 3:length(colnames(REMcRdy))){
combI=cbind.data.frame(combI, REMcRdy[i])
}
Vec1= NA
Vec2= NA
Vec3= NA
Vec4= NA
Vec5= NA
Vec6= NA
Vec7= NA
Vec8= NA
Vec1=NA
Vec2=NA
Vec3=NA
Vec4=NA
Vec5=NA
Vec6=NA
Vec7=NA
Vec8=NA
if(length(REMcRdy)==6){
Vec1=abs(REMcRdy[,3])>=std
Vec2=abs(REMcRdy[,4])>=std
Vec3=abs(REMcRdy[,5])>=std
Vec4=abs(REMcRdy[,6])>=std
bolVec= Vec1 | Vec2 |Vec3 |Vec4
bolVec=Vec1 | Vec2 |Vec3 |Vec4
REMcRdyGT2=REMcRdy[bolVec,1:2]
REMcRdyGT2[ ,3:6]= REMcRdy[bolVec,3:6]
REMcRdyGT2[ ,3:6]=REMcRdy[bolVec,3:6]
shiftOnlyGT2=shiftOnly[bolVec,1:2]
shiftOnlyGT2[ ,3:6]= shiftOnly[bolVec,3:6]
shiftOnlyGT2[ ,3:6]=shiftOnly[bolVec,3:6]
}
if(length(REMcRdy)==8){
@@ -180,11 +181,11 @@ if(length(REMcRdy)==8){
Vec4=abs(REMcRdy[,6])>=std
Vec5=abs(REMcRdy[,7])>=std
Vec6=abs(REMcRdy[,8])>=std
bolVec= Vec1 | Vec2 |Vec3 | Vec4 |Vec5 |Vec6
REMcRdyGT2= REMcRdy[bolVec,1:2]
REMcRdyGT2[ ,3:8]= REMcRdy[bolVec,3:8]
shiftOnlyGT2= shiftOnly[bolVec,1:2]
shiftOnlyGT2[ ,3:8]= shiftOnly[bolVec,3:8]
bolVec=Vec1 | Vec2 |Vec3 | Vec4 |Vec5 |Vec6
REMcRdyGT2=REMcRdy[bolVec,1:2]
REMcRdyGT2[ ,3:8]=REMcRdy[bolVec,3:8]
shiftOnlyGT2=shiftOnly[bolVec,1:2]
shiftOnlyGT2[ ,3:8]=shiftOnly[bolVec,3:8]
}
if(length(REMcRdy)==10){
@@ -196,33 +197,33 @@ if(length(REMcRdy)==10){
Vec6=abs(REMcRdy[,8])>=std
Vec7=abs(REMcRdy[,9])>=std
Vec8=abs(REMcRdy[,10])>=std
bolVec= Vec1 | Vec2 |Vec3 |Vec4|Vec5|Vec6|Vec7|Vec8
REMcRdyGT2= REMcRdy[bolVec,1:2]
REMcRdyGT2[ ,3:10]= REMcRdy[bolVec,3:10]
shiftOnlyGT2= shiftOnly[bolVec,1:2]
shiftOnlyGT2[ ,3:10]= shiftOnly[bolVec,3:10]
bolVec=Vec1 | Vec2 |Vec3 |Vec4|Vec5|Vec6|Vec7|Vec8
REMcRdyGT2=REMcRdy[bolVec,1:2]
REMcRdyGT2[ ,3:10]=REMcRdy[bolVec,3:10]
shiftOnlyGT2=shiftOnly[bolVec,1:2]
shiftOnlyGT2[ ,3:10]=shiftOnly[bolVec,3:10]
}
if(std!=0){
REMcRdy= REMcRdyGT2 # [,2:length(REMcRdyGT2)]
shiftOnly= shiftOnlyGT2 # [,2:length(shiftOnlyGT2)]
REMcRdy=REMcRdyGT2 # [,2:length(REMcRdyGT2)]
shiftOnly=shiftOnlyGT2 # [,2:length(shiftOnlyGT2)]
}
if(std==0){
REMcRdy= REMcRdy # [,2:length(REMcRdy)]
shiftOnly= shiftOnly # [,2:length(shiftOnly)]
REMcRdy=REMcRdy # [,2:length(REMcRdy)]
shiftOnly=shiftOnly # [,2:length(shiftOnly)]
}
# R places hidden "" around the header names. The following
# is intended to remove those quote so that the "" do not blow up the Java REMc.
# Use ,quote=F in the write.csv statement to fix R output file.
#write.csv(combI,file = file.path(outDir,"CombinedKLzscores.csv"),row.names = FALSE)
write.csv(REMcRdy,file=file.path(outDir,"REMcRdy_lm_only.csv"),row.names = FALSE, quote=F)
write.csv(shiftOnly,file=file.path(outDir,"Shift_only.csv"),row.names = FALSE, quote=F)
#LabelStd <- read.table(file= "./parameters.csv",stringsAsFactors = FALSE,sep= ",")
#write.csv(combI,file=file.path(outDir,"CombinedKLzscores.csv"),row.names=FALSE)
write.csv(REMcRdy,file=file.path(outDir,"REMcRdy_lm_only.csv"),row.names=FALSE, quote=F)
write.csv(shiftOnly,file=file.path(outDir,"Shift_only.csv"),row.names=FALSE, quote=F)
#LabelStd <- read.table(file="./parameters.csv",stringsAsFactors=FALSE,sep=",")
LabelStd<- read.csv(file=studyInfo,stringsAsFactors = FALSE)
LabelStd<-read.csv(file=studyInfo,stringsAsFactors=FALSE)
print(std)
LabelStd[,4]= as.numeric(std)
write.csv(LabelStd,file=file.path(outDir,"parameters.csv"),row.names = FALSE)
write.csv(LabelStd,file=studyInfo,row.names = FALSE)
LabelStd[,4]=as.numeric(std)
write.csv(LabelStd,file=file.path(outDir,"parameters.csv"),row.names=FALSE)
write.csv(LabelStd,file=studyInfo,row.names=FALSE)